Research on multi-time scale optimal scheduling of integrated energy system based on digital twinning

Due to the differences in time scales of different units in integrated energy systems, the adaptability of dispatching to load fluctuation is low, and the corresponding cost input is large. Therefore, the research on multi-time scale optimal dispatching of integrated energy systems based on digital twinning is proposed. Based on FPGA and multiple constraints such as timing, resources, and algorithms, a digital twin model of a multi-energy integrated energy system is constructed to address the dynamic changes of various components in the time scale of the multi-energy complementary system. In the scheduling stage, the scheduling interval is determined based on the output and load parameters of each unit module in the integrated energy system, and the specific scheduling parameters are determined with the goal of digital twin mirror equivalence. In the test results, the fluctuation cost under different operation scenarios is stable below 700 yuan, and the operation cost is stable below 1, 400 yuan, both of which are at a low level.


1.Introduction
For the comprehensive energy system, maximizing the effective utilization rate of energy is the key to its economic benefits.However, it is worth noting that the analysis of the dispatching demand of integrated energy systems from the demand side shows that there are obvious differences in the demand for different energy sources in different periods, which is also the key to the waste of resources in most integrated energy systems [1][2] .Given this, it has become one of the most important factors to study the scheduling problem of integrated energy systems from multiple time scales.In the research results of related comprehensive energy system scheduling, some scheduling methods have also achieved good application results.On this basis, a new energy management strategy is proposed, which utilizes the improved BAS algorithm to optimize energy scheduling.The main purpose of this optimization strategy is to improve the economic performance of the comprehensive energy system [3-4]   , gas and heat are taken as the research object, and the objective function constructed is the minimum operating cost, which is taken as the execution basis of the optimal scheduling model.On this basis, the method comprehensively considers the upper and lower limits of the system's electricity, gas, and heat transport capacity, combines the constraints and limitations of the upper and lower limits of the output of various energy conversion equipment on the scheduling, and solves the optimized scheduling model based on the adaptive improvement of the Beetle Antennae Search (BAS) algorithm [5-6]   .In the test results of this method, the reasonable optimal scheduling of the comprehensive energy system can be realized with or without considering the time-sharing energy price, which greatly improves the effective utilization rate of energy.However, it is highly dependent on the reliability of energy demand data, so its application scope has certain limitations [7][8] .On this basis, a new energy utilization method, namely the Sparrow Optimization Algorithm, was proposed and applied to the optimal scheduling of the power system.This method defines the energy sparrow based on the integrated energy system, and after determining the initial population [9][10] , the relative position of the optimal fitness value and the worst fitness value is determined according to the set sparrow optimization principle, which effectively realizes the multi-energy complementarity and conversion processing at the energy level.It has good practical application value in enhancing the reliability of the integrated energy system, but this method lacks consideration of the actual operation of the system, so there is a certain deviation between the effect of the landing stage and the ideal state.
Based on this, this study conducts research on the optimization scheduling problem of multi-time scale comprehensive energy systems based on digital pairing and conducts comparative experimental research.

FPGA-based construction of digital twin image of the comprehensive energy system
This paper first explores the optimal allocation problem of comprehensive energy systems under multiple time scales.The fluctuation of the actual output of different units in an integrated energy system in time scale is fully considered.On this basis, this project has implemented a digital twin design for the energy systems based on FPGA.In the specific design process, the core goal of this paper is to reduce the resource cost of FPGA as much as possible.Therefore, in the development process, three restrictions are observed: timing restriction, resource restriction, and algorithm restriction.Among them, the time sequence restriction means that when designing the digital twin image of the integrated energy system, each unit module in the integrated energy system should be strictly limited in time sequence, to ensure the correctness of the final solution.Given this, this paper adopts the state machine control strategy.Resource limitation means that in an FPGA, resources such as flip-flops, lookup tables, and Digital Signal Processing (DSP), are relatively limited.To quickly solve the parallel state of each unit module in an integrated energy system at different time scales, this paper uses an intellectual property core (IP) composed of multiple modules to improve the number of module reuses and to reduce the scale of resources occupied by each module.Finally, the algorithm is limited.When using differential equations to describe the dynamic process of the integrated energy system, it will also be limited.Given this, this paper solves this problem by taking the data format which is more conducive to ensuring the calculation accuracy as the implementation basis of the digital twin mirror construction of the integrated energy system.
Based on the above, this paper uses the venin theorem to set the same impedance element of each unit module in the integrated energy system to be connected to the same circuit when it is in a steady state, and the corresponding port voltage is in a constant state.It should be pointed out that in a real integrated energy system, the time scale of the mechanical components of the electric motor is longer than that of the power components, but in the extremely short discrete time of electromagnetic simulation, the rotor speed of the two parts presents an approximately constant specificity.This means that at a discrete moment, there will be a transient stable mode.In this paper, the step integration method was used to study the working status of each module during the (n-1, n) period.The specific processing method can be described as follows In this way, the discrete comprehensive energy system state data is processed by a digital twin mirror image, and the obtained form can be expressed as where n m i is a digital twin diagram that reflects the operational status of the energy integration system.
On this basis, a digital twin model of a multi-energy complementary system is established to provide an implementation basis for optimal scheduling at multiple time scales.

Integrated energy system optimization scheduling
Based on this model, this project will use the output and load parameters of each unit module of the multi-energy system as the basis, and modify the scheduling intensity through digital twin images of the operation status data of the multi-energy system.
When the load parameter is Q, the initial scheduling parameters output by each unit module in the integrated energy system can be expressed as 0 ( ) with the help of digital twin mirror images can be expressed as 1 ( ) ( ) In this way, we determine the final () qx  value, it should be noted that at this time, () qx  is necessary to meet the interval range shown in Equation ( 3), and when it cannot be met, it is necessary to schedule the energy of the energy storage device.

Test preparation
When analyzing the practical application effect of the multi-time scale optimal scheduling method of the integrated energy system designed in this paper, based on an actual hybrid renewable energy power system, a comparative test study is carried out in the simulation environment.In the specific test process, the simulation test environment includes five main parts: WT, PV alkaline electrolytic cell, hydrogen storage tank, hydrogen gas turbine, and energy storage battery.On this basis, the specific parameter information of different components is set, and the relevant data information is shown in Table 1.On this basis, eight experimental scenarios were constructed to verify the effectiveness of multi-time scale optimization scheduling for the developed multi-time scale integrated energy system.In terms of experimental research, this project will take the improved sparrow optimization algorithm and the improved BAS algorithm as the research objects, to construct a power grid comprehensive scheduling model with energy integration as the research object.By comparing the test results under different scheduling methods, the application value of the scheduling method designed in this paper is objectively evaluated.

Test results and analysis
Based on the above test environment, three scheduling methods are used to carry out comparative tests, and the cost input is taken as a specific evaluation index.The data obtained are shown in Table 2.As shown in Table 2, there is a significant difference between the fluctuation cost and operating cost when using three optimal control strategies.Among them, when the peak value and valley value of comprehensive load are fixed (test scenario 1-test scenario 4, test scenario 5-test scenario 8), the fluctuation cost and operation cost of the three scheduling methods are relatively stable.However, the fluctuation cost and operation cost of the integrated energy system optimal scheduling method based on the improved sparrow optimization algorithm are higher than those of the integrated energy system optimal scheduling method based on the improved BAS algorithm and the scheduling method designed in this paper under high load fluctuation (test scenario 5 -test scenario 8).When the load fluctuation is low (test scenario 1 -test scenario 4), the fluctuation cost and operation cost of the integrated energy system optimal scheduling method based on the improved BAS algorithm are higher.On this basis, the fluctuation cost and operation cost of different comprehensive load peak and valley states are compared.Among them, the fluctuation cost and operation cost of the integrated energy system optimal dispatching method based on the improved sparrow optimization algorithm change greatly, reaching an increase of about 200.0 yuan, while the integrated energy system optimal dispatching method based on the improved BAS algorithm is relatively stable.At that time, the overall fluctuation cost and operation cost were always at a relatively high level (the fluctuation cost was in the range of 800 yuan -900 yuan and the operation cost was 14, 000 yuan -15, 000 yuan).In contrast, in the test results of the scheduling method designed in this paper, not only do the fluctuation cost and operation cost under different load conditions always maintain high stability, but also the specific cost parameters are relatively low, among which the fluctuation cost is stable below 700 yuan and the operation cost is stable below 1400 yuan.

Conclusion
In this paper, the research on multi-time scale optimal scheduling method of integrated energy systems based on digital twinning is put forward.From the perspective of the composition of the integrated energy system and load demand, the energy scheduling problem is analyzed with the help of digital twinning technology, and the specific scheduling is carried out in combination with the actual output and load state of each unit of the integrated energy system in different periods.With the help of the dispatching method designed in this paper, the cost of the integrated energy system in different operating states is effectively reduced, which has good practical application value for improving the economic benefits of the integrated energy system.
where 1 n m hist − represents the discretization result of the operation state of each unit module in the integrated energy system, m L is the excitation inductance parameter indicating that the same impedance element is connected to the same circuit, t  represents the time difference between the mechanical and electrical components of the motor, n m v represents the rotational speed of the rotor at time n, and 1 n m v − represents the rotational speed at time n-1.

L
) qx  represents the internal of the integrated energy system x scheduling parameters of unit modules, express x peak value and valley value output by unit module respectively, and m C represents the steady-state equivalent impedance of the integrated energy system.On this basis, further calibration of () qx 

Table 1 .
Parameter Configuration Information of Testing Integrated Energy System.

Table 2 .
Comparison Table/Element of Test Results of Different Scheduling Methods.