Research and Optimization of Aircraft Remaining Oil Control System based on Fuzzy Algorithm

Aiming at the poor accuracy and stability of the existing aircraft remaining oil control system, an amount of oil remaining system for aircraft based on a fuzzy algorithm is proposed based on capacitance detection and mechanical detection, with the introduction of fuzzy algorithm optimized PID control technology. The system can form flight variables control system according to the error between the actual remaining oil and the ideal remaining oil and realize the stable control of the aircraft remaining oil. Furthermore, through targeted case analysis, the deviation of aircraft residual fuel before and after optimization is compared so as to further verify the reliability and feasibility of the designed aircraft residual fuel control system based on a fuzzy algorithm. In conclusion, the fuzzy algorithm is efficient in the optimization of the aircraft remaining oil control system.


Introduction
With the structural upgrading of China's aviation industry, and in order to comply with the development trend of China's aviation industry as well as to create an advanced and efficient aircraft production system, aircraft manufacturing enterprises are paying more attention to the research and development and upgrading of aircraft equipment year by year [1].The remaining oil quantity is one of the main indexes of aircraft flight safety.To some extent, it can be said that the remaining oil quantity is related to whether an aircraft can land safely or not [2].The control of aircraft fuel consumption rate is an important task because it is affected by many factors such as flight speed, altitude, fuel supply system, and wing resistance.In addition to taking measures such as simulation calculation, observing fuel consumption process, reading data, and controlling flight duration, advanced and efficient aircraft fuel systems are under construction [3].Each type of aircraft trial base pays more attention to this aspect of research.In order to improve the level of flight safety, the technical innovation of aircraft fuel systems is constantly carried out, focusing on improving the intelligent and automatic level of flight control systems [4].As an important index of aircraft flight safety, the redundancy of the aircraft remaining oil weight has a great impact on the final safe return.Therefore, the control accuracy and stability of the existing remaining oil system need to be improved [5].In this paper, based on the existing relatively mature capacitance detection technology, a capacitance sensor is used as the oil amount detection system to identify the remaining oil number of aircraft.By introducing PID control technology, taking the mass error between real-time oil quantity and expected value as the control quantity into consideration, as well as using a fuzzy control algorithm and corresponding fuzzy rules to adjust the controller parameters, the traditional PID controller is further optimized and designed [6].A set of aircraft remaining oil control systems based on a fuzzy algorithm is designed to realize the stable control of remaining oil.

Oil Control System Structure
There are a large number of fuel volume control systems in China, such as structure and function.According to the functions realized by each module, the oil volume control system is roughly composed of an oil volume detection system, speed control module, altitude control module, flight control system, and other major functional modules [7].The functions of each module are as follows:  Oil quantity detection system: when the aircraft needs to detect the remaining oil quantity in the tank, the capacitor in the module will detect the sensor to it and produce detection signals for the remaining oil quantity in the mailbox. Speed control module: the engine power of the aircraft is adjusted accordingly to change the flight speed and fuel consumption of the aircraft. Altitude control module: the angle of attack of the aircraft and the height of the wing is adjusted.Due to different atmospheric oxygen content and resistance, the fuel consumption rate is affected. Flight control system: a combination of automatic flight control subsystems of various functions on an aircraft provides overall control of altitude, speed, and lateral track.The structure of the remaining oil quantity control system is shown in Figure 1.

Oil Quantity PID Control System
As a common control method, PID control is widely used in automatic control systems.PID controllers have the advantages of simple structure, relatively sensitive adjustment, and can eliminate steady-state errors [8].They also have good control effects for some control objects, such as linear systems.On the basis of traditional fuel control systems, this paper introduces the PID control concept, obtains the remaining fuel data of the aircraft through the fuel detection system, and compares it with the set value.The oil amount error between the actual remaining number of aircraft and the set value is used as the control amount to stably control the remaining amount of aircraft oil.

PID Controller Parameters
The PID controller actually adjusts the controlled object through three parallel structures: proportional, integral, and differential.Furthermore, the PID controller normally takes the deviation value as the input and outputs the control quantity.
The proportion link in the controller can amplify the deviation value according to a certain proportion.When the feedback value is not equal to the set value, the proportion link will execute the amplification command, thereby reducing the deviation value.Generally speaking, increasing the scale factor can accelerate the response speed of the system, but it is also more likely to cause static errors and oscillations in the system.The integration link in the controller can reduce and eliminate the steady-state error of the system.On the other hand, the introduction of the integration link may slow down the adjustment speed of the system.
The differential link in the controller can further reduce the system's adjustment time and adjust the magnitude of the differential constant, which can accelerate the system's adjustment speed and improve the system's ability to suppress vibration.Still, it may also reduce the system's resistance to high-frequency signals.

Fuzzy Control Theory
Fuzzy control is an advanced modern intelligent control method developed based on fuzzy set theory, fuzzy language, fuzzy mathematics, and fuzzy logic inference as theoretical references and theoretical foundations, and it has been developed well based on relevant expert research and a large number of engineering practices [9].It has the characteristics of low cost and simple utilization method and is suitable for most automatic control systems [10].
When working with fuzzy PID, the control rules used by the fuzzy controller will be formulated based on the engineering experience of relevant researchers and the laws in a large number of engineering practices [11].This is the most critical part of this control method.After that, the functional relationship corresponding to the control method will be found, and the fuzzy output value will be obtained.Finally, the fuzzy resolution link will be implemented [12].
The use of different controllers, different topologies, different fuzzy decision rules, and different settings for the amplification coefficient of fuzzy inputs and outputs in fuzzy controllers will directly affect the control effect of the fuzzy PID control method.

Composition of Fuzzy Controller
As the core mechanism of the entire fuzzy control, the fuzzy controller is mainly composed of four parts.
1) Input fuzzy: Input fuzzy mainly refers to mapping the detected input physical quantities to the fuzzy universe in an appropriate proportion.This mapping can ultimately achieve the goal of blurring the input quantities by using simplified variables to express the detected inputs.
2) Knowledge base: The database and rule base form a knowledge base in parallel, where the database can be used to process the fuzzy input quantity obtained after inputting fuzzy, and the rule base can achieve the control function of the controller.By writing different control rules, different control effects can be obtained, and differences in control rules can significantly affect the final performance of the controller.Generally, the Fuzzy control box in MATLAB can be used to write the rule base, and the writing language is often "if/or then".
3) Fuzzy inference: Fuzzy inference generally uses the machine language of a fuzzy controller to simulate humans based on the implication relationships and fuzzy control rules that should exist in the logic of the actual control process so as to analyze and judge the input fuzzy control quantities and infer the fuzzy decision output which needs to be resolved.
4) Anti-fuzziness: Due to the output quantity obtained through fuzzy reasoning and accurate control quantity, the fuzzy controller required in actual production and life has precision requirements for its output control quantity.Therefore, it is necessary to accurately process the output quantity after fuzzy reasoning through the anti-fuzziness step.Generally, precision algorithms such as the maximum membership method can be used to convert the fuzzy output into a clear quantity within the scope of the universe and then convert it into the actual required control quantity through scale transformation.
The composition diagram of the fuzzy controller used in this article is shown in Figure 2.

Membership Function Establishment of PID Controller
During the fuzzification process of the input value, the range of input value after fuzzification is {NB, NM, NS, ZO, PS, PM, PB}, which has the following meanings: negative big, negative middle, negative small, zero, positive small, positive middle, and positive big.Meanwhile, the range of fuzzy subsets is specified as [-6,6], and can be adjusted through input by proportional parameters.The final output obtained after fuzzification is . As a parameter regulator for PID, it can both improve current controller parameters and continuously update them online over time.Similarly, the subset of fuzzy outputs before fuzzification is also {NB, NM, NS, ZO, PS, PM, PB}, and its range is also defined as [-6,6].At the same time, the output can be adjusted by proportional parameters.Proportional parameters are also known as quantization factors, which are adjusted appropriately based on the actual control system needs to achieve better control results.However, before and after the change, the fuzzy universe still corresponds to the fuzzy words {NB, NM, NS, ZO, PS, PM, PB} used by the seven algorithms.
In the fuzzy controller used in this paper, the input term is 2, the output term is 3, and its topology is shown in Figure 3.

Figure 3. Fuzzy controller input output topology
Here in this paper, the input and output membership functions of the fuzzy PID controller are all subject to triangular membership functions.In the corresponding coordinate system, the horizontal axis coordinate is a fuzzy domain, and the vertical axis membership function value and the vertical axis coordinate only have mathematical significance practical physical significance, so they all have no units.The distribution of the triangular membership function curve here is shown in Figure 4. can be performed to obtain proportional, integral, and differential parameters.At a special moment, these newly obtained parameters will be more consistent with the system's operation, and this very consistency can keep the weight automatic control system in quite an ideal control state.As a result, the data input quality will be improved greatly.

Back-fuzzy Modeling
Due to the need for accurate deterministic values as control outputs in practical applications as well as in order to obtain a truly accurate output as the input for the next step, the fuzzy output of the controller needs to be deblurred.In this paper, the gravity-center method is selected to achieve this.The expression of the center of gravity method is as follows: where 0 z is the precise output obtained after ambiguity resolution, i z is the value in the universe of fuzzy control quantities, and ) ( i c z u is the membership value of i z .

Experimental Analysis
In order to evaluate the control effect of the oil quantity control system based on a fuzzy algorithm, a specific experiment is designed for testing and analysis.Two flight simulation devices of the same model were selected for comparison and verification.One device completed the optimization of the control system, while the other did not.Two flight simulation devices of the same model were selected for comparison and verification.One device completed the optimization of the control system, while the other did not.As can be seen from Figure 5, before the optimization of the control system, the weight of each component fluctuated greatly.After optimizing the control system, the remaining oil can be controlled better.

Conclusion
In this paper, the fuzzy control algorithm is used to optimize the three parameters of PID control in real time, and the method is applied to the aircraft remaining oil control system.The internal module and its corresponding function are defined.The fuzzy control principle and PID control method of the control system are briefly introduced.Based on expert experience and through using corresponding fuzzy control rules and membership functions, a fuzzy controller is designed to optimize PID parameters using the quality deviation value between the real-time product and the set value as the control variable.Through the targeted simulation analysis, the control system before and after optimization is tested.Compared with the traditional aircraft remaining oil control system, the proposed remaining oil control system based on a fuzzy algorithm has higher control accuracy and better anti-interference performance.

Figure 1 .
Figure 1.Structure diagram of oil volume control system

Figure 2 .K
Figure 2. Fuzzy controller composition diagram4.Fuzzy PID Controller Design StepsBased on the theory mentioned above, the fuzzy PID controller can be designed in the following steps:4.1.Controller Parameter SettingBy comparing the real-time detection value of the component weight with the set ideal weight and obtaining the difference, the obtained input value is input into the fuzzy controller, followed by fuzzy processing, fuzzy rule mapping, logical reasoning judgment, and precision of the fuzzy data.Proportional, integral, and differential controller parameters can be obtained separately, and finally, these three parameters can be adjusted online in real-time.

Figure 4 ..
Figure 4. Curve distribution of triangular membership function 4.3.Fuzzy Rule Making Fuzzy rules can be embodied in the form of fuzzy rule tables.The establishment of fuzzy rule tables mainly follows the relationship between the three parameters of the PID controller and the control effect in actual work.Based on experts' long-term experience in system debugging and continuous debugging of performance parameters, the fuzzy rule tables are summarized and refined as P K  ,

Figure 5 .
Figure 5.Comparison before and after optimization