Optimized regulation and control method for multi-type energy storage and adjustable load based on improved model predictive control and related device

Abstract

The invention provides a multi-type energy storage and adjustable load optimization regulation and control method based on improved model predictive control and a related device. The method comprises the following steps: acquiring a dynamic characteristic model of multi-type energy storage and adjustable load; based on the parameters provided by the dynamic characteristic model, performing rolling solution on the pre-constructed hierarchical predictive control model, and forming a corresponding control instruction; and dividing the control instruction according to the time scale, and outputting the control instruction to the corresponding multi-type energy storage and adjustable load for execution. According to the method, the dynamic characteristic model of the multi-type energy storage and the adjustable load is obtained, so that the problems of heavy calculation burden and high model mismatching risk when a high-dimensional and nonlinear system is processed are reduced, and refined coordinated regulation and control of resources are realized. Meanwhile, the control instructions are divided according to the time scale and executed according to the power distribution priority, the system operation cost is reduced, the regulation and control efficiency is improved, the resource potential is deeply mined, and the full life cycle cost of the energy storage equipment and the adjustable load user experience are comprehensively considered.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy system regulation and control, and specifically relates to an optimization regulation method and related devices for multiple types of energy storage and adjustable loads based on improved model predictive control. Background Art

[0002] As the global energy mix accelerates toward a clean, low-carbon transition, large-scale grid integration of renewable energy sources, such as wind power and photovoltaics, has become a key development direction in the energy sector. This integration approach brings significant environmental and economic benefits to the energy system, helping to reduce reliance on traditional fossil fuels, lower carbon emissions, and promote sustainable development.

[0003] However, the intermittent and fluctuating output of renewable energy presents numerous challenges to grid operations, such as increased difficulty balancing power and increased peak load regulation. Meanwhile, energy storage technologies are diversifying, with the emergence of various types, including lithium batteries, flow batteries, and hydrogen energy storage. Adjustable load resources are also being deployed on a large scale, such as industrial flexible loads, electric vehicle clusters, and temperature-controlled loads in smart buildings. These opportunities present new opportunities for enhancing energy system flexibility and strengthening the grid's capacity to absorb energy.

[0004] Current research on the optimization and control of multiple types of energy storage and adjustable loads primarily employs traditional optimization methods and Model Predictive Control (MPC) technology. Traditional optimization methods, which solve for optimal control strategies by establishing deterministic mathematical models, struggle to cope with the uncertainties and complex operational constraints of renewable energy. Traditional MPC suffers from heavy computational burdens and high risk of model mismatch when dealing with high-dimensional, nonlinear systems. Furthermore, existing research often overlooks the differences in the characteristics of different types of energy storage and load resources, making it impossible to achieve refined coordinated control of resources. Furthermore, with the increasing complexity of energy systems, traditional optimization and control methods cannot simultaneously meet both real-time and optimization accuracy requirements when dealing with the coordinated optimization of multiple types of energy storage and adjustable loads. This results in increased system operating costs and reduced control efficiency. Furthermore, there is a lack of comprehensive consideration of the full lifecycle costs of energy storage equipment and the user experience of adjustable loads, limiting the in-depth exploration of resource potential. Summary of the Invention

[0005] In view of this, the present invention provides an optimization control method and related devices for multiple types of energy storage and adjustable loads based on improved model predictive control, aiming to achieve efficient and coordinated utilization of multiple types of energy storage and adjustable load resources, and promote the development of energy systems towards intelligence and low carbon.

[0006] In order to achieve the above object, the technical solution provided by the present invention is as follows:

[0007] In a first aspect, the present invention provides a method for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control, comprising the following steps:

[0008] Obtain dynamic characteristic models of multiple types of energy storage and adjustable loads;

[0009] Based on the parameters provided by the dynamic characteristic model, the pre-built hierarchical predictive control model is solved in a rolling manner and the corresponding control instructions are generated;

[0010] Divide control instructions by time scale and output them to corresponding multi-type energy storage and adjustable loads for execution;

[0011] Among them, the hierarchical predictive control model is used to model the preset optimization objectives and stipulate the power allocation priority. The power allocation priority is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales.

[0012] Furthermore, the objective function of the hierarchical predictive control model is determined by the sum of the products of several optimization objectives and corresponding weights; the weights include at least stability weights and economic weights; the stability weights are used to measure the importance of the optimization objectives of ensuring grid stability in the objective function, and the economic weights are used to reflect the importance of the optimization objectives focusing on economy in the objective function; the stability weights increase with the increase of the rate of change of new energy power, and the economic weights increase with the increase of the degree to which the average SOC of energy storage deviates from the set value and / or the SOC difference between energy storage units.

[0013] Furthermore, if the optimization objectives include grid stability optimization objective, economic optimization objective, SOC balance optimization objective, and SOC variance minimization optimization objective, the corresponding weights are determined according to the following formulas:

[0014]

[0015]

[0016]

[0017]

[0018] Where, is the weight of the grid stability optimization objective, which belongs to the stability weight. To contribute to new energy, t is time, is a constant; It is the weight of the economic optimization goal, which belongs to the economic weight. It is a quantitative value of the degree to which the average SOC of the energy storage deviates from the set value; It is the weight of the SOC balance optimization target, which belongs to the economic weight. is the quantified value of the maximum SOC difference between energy storage units; The weight of the optimization objective for minimizing the SOC variance is an economic weight; 、 、 and are the preset coefficients of each weight respectively.

[0019] Furthermore, the constraints of the hierarchical predictive control model include energy storage power constraints, SOC boundary constraints, and load regulation constraints. The energy storage power constraints, SOC boundary constraints, and load regulation constraints are all processed using slack variables, and penalty terms are used to limit each constraint. The constraints and penalty terms after relaxation are as follows:

[0020] Energy storage power constraint relaxation:

[0021]

[0022] Where, is the power of electrochemical energy storage at time k, is the power relaxation variable of electrochemical energy storage, and are the lower and upper limits of the power of electrochemical energy storage, respectively;

[0023] SOC boundary constraint relaxation:

[0024]

[0025] Where, is the energy storage state of charge at time k, is the slack variable of the SOC constraint, and are the lower and upper limits of the energy storage state of charge, respectively;

[0026] Load regulation constraint relaxation:

[0027]

[0028] Where, is the adjustable power of the adjustable load at time k, is the power slack variable of the adjustable load, and They are the lower and upper limits of the adjustable power of the adjustable load respectively;

[0029] Penalty items:

[0030]

[0031] Where, is the penalty term, N is the total number of calculation periods, 、 、 They are 、 、 The penalty coefficient;

[0032] The penalty coefficient is determined as follows:

[0033]

[0034] Where, is the standard deviation of the forecast error of renewable energy output, is the basic penalty coefficient, To adjust the gain factor.

[0035] Furthermore, the objective function of the hierarchical predictive control model is as follows:

[0036]

[0037]

[0038]

[0039]

[0040]

[0041] Where, 、 、 and They are the grid stability optimization objective function, the economic optimization objective function, the SOC balance optimization objective function and the SOC variance minimization optimization objective function, respectively. 、 、 and They are 、 、 and The weight of; N is the total number of calculation periods, is the actual power of the power grid at time k, is the corrected reference power trajectory; 、 and They are energy storage operation cost, adjustable load compensation cost and new energy power abandonment cost. 、 and They are 、 and The weight coefficient; M is the total number of energy storage units, is the SOC value of the i-th energy storage unit at time k, is the average SOC value at time k.

[0042] Furthermore, in the rolling solution of the pre-built hierarchical predictive control model, error feedback is used to correct the output of new energy, and in the SOC balance optimization process, SOC dynamic balance is used to adjust the power distribution priority of each energy storage unit. The error feedback correction and SOC dynamic balance are as follows:

[0043] The error feedback correction is as follows:

[0044]

[0045]

[0046] Where, is the actual output of new energy at time k, is the predicted output of new energy at time k, is the k-time error, is the error attenuation coefficient; Contribute to the revised new energy;

[0047] The dynamic balance of SOC is as follows:

[0048] The initial power distribution weight of each energy storage unit is calculated as follows:

[0049]

[0050] Where, Assign a weight to the initial power of the i-th energy storage unit, is the SOC deviation of the i-th energy storage unit, is the SOC deviation of the j-th energy storage unit;

[0051] The corrected power distribution weights after SOC dynamic balancing are as follows:

[0052]

[0053] Where, Assign a weight to the corrected power of the i-th energy storage unit, and are the lower and upper limits of the SOC safe operating range.

[0054] Furthermore, for electrochemical energy storage, flywheel energy storage, and adjustable loads, the power allocation priorities are as follows:

[0055] The time scale from seconds to minutes is allocated to flywheel energy storage; the time scale from minutes to hours is allocated to electrochemical energy storage; and the time scale above hours is allocated to adjustable loads.

[0056] In a second aspect, the present invention provides an optimized control device for multiple types of energy storage and adjustable loads based on improved model predictive control, comprising:

[0057] Model acquisition module, used to obtain dynamic characteristic models of multiple types of energy storage and adjustable loads;

[0058] The model solving module is used to perform rolling solution on the pre-built hierarchical predictive control model based on the parameters provided by the dynamic characteristic model and generate corresponding control instructions;

[0059] An optimization and control module is used to divide control instructions by time scale and output them to the corresponding multi-type energy storage and adjustable loads for execution;

[0060] Among them, the hierarchical predictive control model is used to model the preset optimization objectives and stipulate the power allocation priority. The power allocation priority is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales.

[0061] In a third aspect, the present invention provides a computer device, comprising a processor and a memory:

[0062] The memory is used to store computer programs and send instructions of the computer programs to the processor;

[0063] The processor executes the optimization control method of multiple types of energy storage and adjustable loads based on improved model predictive control according to the instructions of the computer program as described in the first aspect.

[0064] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the present invention implements a method for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control as in the first aspect.

[0065] In summary, the present invention provides an optimization control method and related devices for multiple types of energy storage and adjustable loads based on improved model predictive control. The method obtains dynamic characteristic models of multiple types of energy storage and adjustable loads; based on the parameters provided by the dynamic characteristic models, a pre-built hierarchical predictive control model is rolled and solved to form corresponding control instructions; the control instructions are divided according to the time scale and output to the corresponding multiple types of energy storage and adjustable loads for execution; wherein the hierarchical predictive control model is used to model the preset optimization target and specifies the power allocation priority, which is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales. The present invention avoids the problem of ignoring the differences in resource characteristics in existing studies by obtaining the dynamic characteristic models of multiple types of energy storage and adjustable loads, and rolls and solves the hierarchical predictive control model based on this. Compared with traditional MPC, it reduces the problems of heavy computational burden and high risk of model mismatch when processing high-dimensional and nonlinear systems, and realizes refined collaborative control of resources. At the same time, control instructions are divided according to time scales and executed according to power allocation priorities, overcoming the defects of traditional optimization methods that are difficult to cope with the uncertainty of renewable energy and complex operating constraints, breaking the limitation of a single optimization algorithm that cannot take into account both real-time performance and optimization accuracy, reducing system operating costs, improving control efficiency, deeply tapping resource potential and comprehensively considering the full life cycle cost of energy storage equipment and the user experience of adjustable loads. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0067] Figure 1 A flowchart of a method for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control provided by an embodiment of the present invention;

[0068] Figure 2 A calculation flow chart of a method for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control provided by an embodiment of the present invention;

[0069] Figure 3 A block diagram of a device for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control provided by an embodiment of the present invention;

[0070] Figure 4 A block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purposes, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0072] See also Figure 1 The embodiment of the present invention provides an optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control, comprising the following steps:

[0073] S1: Obtain dynamic characteristic models of multiple types of energy storage and adjustable loads.

[0074] It should be noted that multi-type energy storage refers to energy storage devices with different physical principles or technical routes, such as electrochemical energy storage (such as lithium batteries, flow batteries), mechanical energy storage (such as flywheel energy storage, compressed air energy storage) and electromagnetic energy storage (such as supercapacitors).

[0075] Adjustable load refers to load equipment whose power consumption can be adjusted according to dispatch instructions, such as flexible industrial loads (such as adjustable speed motors, heat pumps), electric vehicle charging loads, energy storage air conditioners, etc., and its power consumption characteristics have a certain degree of flexibility.

[0076] The dynamic characteristics model is a mathematical model that describes the electrical and physical characteristics of energy storage and loads in dynamic processes (such as charging and discharging, and power regulation). It is usually expressed using state-space equations, transfer functions, or differential equations to reflect its power response speed, capacity limitations, efficiency characteristics, etc.

[0077] This step involves mechanistic analysis to determine the dynamic parameters of the energy storage and load under different operating conditions (such as time constants, capacity decay coefficients, and power regulation rates), thereby establishing mathematical models that accurately characterize their dynamic behavior. For example, lithium batteries can use the Thevenin equivalent circuit model to describe the dynamic relationship between terminal voltage and charge and discharge current, while flywheel energy storage can be modeled using the dynamic equations of moment of inertia and torque. These models form the foundation for subsequent control algorithm design, predicting the system's response to different control commands.

[0078] S2: Based on the parameters provided by the dynamic characteristics model, a pre-built hierarchical predictive control model is solved in a rolling manner to generate corresponding control instructions. The hierarchical predictive control model is used to model the preset optimization objectives and define power allocation priorities. The power allocation priorities are used to output corresponding control instructions to the corresponding energy storage or adjustable loads at different time scales.

[0079] It should be noted that the hierarchical predictive control model decomposes the control problem into a multi-layer optimization structure with different time scales or priorities. Rolling solve is the core mechanism of model predictive control (MPC). Within each control cycle, the optimization problem is solved based on the current system state and future forecast data. Only the control instructions at the current moment are executed, and the process is repeated in the next cycle, forming a closed-loop control.

[0080] This step decomposes the complex optimization problem into multiple subproblems using a hierarchical structure, reducing computational complexity. Each subproblem is coupled through optimization objectives and constraints. A dynamic characteristic model is then used to predict the system's future state (such as the energy storage SOC and load power demand). Within each control cycle, an optimization problem (e.g., quadratic programming) is solved, including constraints (such as energy storage capacity limits and load power bounds), to obtain the optimal control instructions for the current moment. This method offers real-time adaptability and can address uncertainties such as fluctuations in renewable energy output.

[0081] S3: Divide the control instructions by time scale and output them to the corresponding multi-type energy storage and adjustable loads for execution.

[0082] It should be noted that time-scale division refers to the division of control commands into different frequency bands based on their response speed requirements. Power allocation priorities are pre-defined resource scheduling sequences, such as prioritizing the use of fast-response energy storage (such as flywheels) to handle high-frequency fluctuations, followed by the use of adjustable loads (such as flexible industrial loads) to offset low-frequency power demands.

[0083] This step matches control instructions to corresponding devices on a time scale based on the dynamic characteristics of energy storage and load (such as response speed and capacity). Through priority allocation, hierarchical regulation is achieved, with "fast devices handling high-frequency fluctuations and slow devices handling low-frequency demands", optimizing resource utilization efficiency and extending equipment life.

[0084] This embodiment provides an optimization and control method for multiple types of energy storage and adjustable loads based on improved model predictive control. By deeply integrating the dynamic characteristic model into the improved hierarchical model predictive control framework, the optimization objective is decoupled into long-term optimization scheduling and short-term power tracking through time scale stratification. Based on the differences in the dynamic characteristics of energy storage and load, a multi-time scale power allocation priority rule is constructed. The optimization problem is solved in real time within each control cycle, and the system state prediction is updated in real time to achieve refined control of "rapid response to high-frequency fluctuations and capacity matching of low-frequency demand". It effectively resolves the contradiction between equipment capacity and response speed in traditional control, significantly improves the system's adaptability and control accuracy to uncertain factors such as renewable energy output and load demand, and provides an efficient technical path for the intelligent control of integrated energy systems.

[0085] See also Figure 2 , Figure 2 The following describes the computational flow of a method for optimizing the control of multiple types of energy storage and adjustable loads based on improved model predictive control. The following describes the dynamic characteristic models of multiple types of resources, including electrochemical energy storage, flywheel energy storage, and adjustable loads, to facilitate the introduction of subsequent examples.

[0086] (1) Electrochemical energy storage

[0087] The equivalent circuit model of the electrochemical energy storage model uses a second-order RC network to characterize the battery polarization effect, and the dynamic equation is:

[0088]

[0089] The constraints are: , .

[0090] In the formula, SOC represents the state of charge of the battery, represents the rate of change of state of charge, Indicates the battery charge and discharge current, Indicates the rated capacity of the battery; Represents the terminal voltage of the battery, represents the open circuit voltage of the battery, Represents the internal resistance of the battery, and represents the voltage across the two RC networks in the battery equivalent circuit, and Corresponding to the resistance of the two RC networks, and Capacitors corresponding to the two RC networks; SOC max and SOC min Indicates the upper and lower limits of the state of charge, and Indicates the minimum and maximum current allowed by the battery;

[0091] (2) Flywheel energy storage

[0092] Mechanical-electrical coupling equations of the flywheel energy storage model:

[0093]

[0094] in, is the angular velocity of the flywheel, represents the power of the flywheel energy storage system, represents the moment of inertia of the flywheel, represents the loss coefficient of the flywheel, is the power command switching function, For conversion efficiency.

[0095] The speed constraints are: , and Indicates the upper and lower limits of the speed constraint.

[0096] (3) Adjustable load

[0097] Piecewise linearization model of adjustable load model: For loads such as industrial motors and air conditioners, a piecewise function of power regulation and delayed response is defined:

[0098]

[0099] in, It represents the power of the adjustable load at time t, N is the number of segments, is a unit step function, is the adjustment coefficient, is the time-sharing delay parameter, and is the minimum delay time and the maximum delay time.

[0100] The key parameter differences between the energy storage and adjustable loads are quantified as follows:

[0101] The power regulation rate (α) is quantified by the ratio of the power adjustable range to the rated power per unit time:

[0102]

[0103] Where, and Respectively and The power of the moment, Indicates the rated power of the load, Indicates a time interval;

[0104] The capacity boundaries are quantified by the duration and energy limits of the adjustable power:

[0105]

[0106] Where, Represents the capacity boundary, Indicates the starting time, Indicates the length of the time interval, and Respectively represent the lower limit and upper limit of the adjustable power;

[0107] The response lag is the time delay from the control command to the actual power response. The transfer function is fitted through the step response experiment:

[0108]

[0109] Where, represents the transfer function, represents the Laplace transform form of the actual power response in the complex frequency domain, is the Laplace transform form of the control instruction in the complex frequency domain, is the transfer function form of the delay link.

[0110] In one embodiment of the present invention, the objective function of the hierarchical predictive control model is determined by the sum of the products of several optimization objectives and corresponding weights; the weights include at least stability weights and economic weights; the stability weights are used to measure the importance of the optimization objectives of ensuring grid stability in the objective function, and the economic weights are used to reflect the importance of the optimization objectives focusing on economy in the objective function; the stability weights increase with the increase of the rate of change of new energy power, and the economic weights increase with the increase of the degree to which the average SOC of energy storage deviates from the set value and / or the increase of the SOC difference between energy storage units.

[0111] In this embodiment, the ratio of stability weight to economic weight in the objective function is dynamically adjusted based on real-time measurements of renewable energy output fluctuations. This dynamic weight allocation mechanism aims to dynamically adjust the weight coefficients of each sub-objective in a multi-objective optimization problem based on real-time system conditions (such as renewable energy power fluctuations, energy storage SOC levels, and load scalability potential), thereby achieving adaptive optimization of the control strategy. During periods of significant fluctuation, grid stability is prioritized, while during periods of stability, economic efficiency and equipment lifespan are prioritized.

[0112] The stability weight is used to measure the importance of ensuring grid stability in the objective function, and it is dynamically adjusted based on the real-time detection value of the fluctuation range of renewable energy output. The larger the value, the more drastic the fluctuation of renewable energy output. The higher it is, the more the system will prioritize grid stability.

[0113] In a further embodiment of the present invention, if the optimization objectives include a grid stability optimization objective, an economic optimization objective, an SOC balance optimization objective, and an SOC variance minimization optimization objective, the corresponding weights are determined according to the following formulas:

[0114] (1) Weight of grid stability optimization objective:

[0115]

[0116] Where, is the weight of the grid stability optimization objective, which belongs to the stability weight. To contribute to new energy, t is time, is a constant;

[0117] The economic weight reflects the importance of economic performance (reducing operating costs, etc.) in the objective function. Its adjustment is based on two aspects: one is the degree to which the average SOC of the energy storage deviates from the set value. , when the average SOC of energy storage deviates from the set value, the economic weight is increased to reduce the operating cost; secondly, the SOC difference between energy storage units .

[0118] (2) Weight of economic optimization objective:

[0119]

[0120] Where, It is the weight of the economic optimization goal, which belongs to the economic weight. It is a quantitative value of the degree to which the average SOC of the energy storage deviates from the set value;

[0121] (3) Weight of grid stability optimization objective:

[0122]

[0123] Where, It is the weight of the SOC balance optimization target, which belongs to the economic weight. is the quantified value of the maximum SOC difference between energy storage units; The larger the value is (i.e. the greater the SOC difference between energy storage units), the higher the relevant value of the economic weight of this part.

[0124] (4) Weight of grid stability optimization objective:

[0125]

[0126] Where, The weight of the optimization objective for minimizing the SOC variance is an economic weight; 、 、 and are the preset coefficients of each weight respectively.

[0127] In a further embodiment of the present invention, the constraints of the hierarchical predictive control model include energy storage power constraints, SOC boundary constraints and load regulation constraints. The energy storage power constraints, SOC boundary constraints and load regulation constraints are all processed using slack variables, and penalty terms are used to limit each constraint. By introducing slack variables, the original strict hard constraints are transformed into constraints with a certain degree of flexibility. When the power of new energy fluctuates violently or the prediction error is large, the constraints are allowed to be relaxed in a short time to ensure that the optimization problem can be solved. At the same time, by setting a penalty term to limit the amount of slack, excessive deviation from the original constraints is avoided, and system stability is guaranteed.

[0128] The constraints and penalties after relaxation are as follows:

[0129] (1) Relaxation of energy storage power constraints:

[0130]

[0131] Where, is the power of electrochemical energy storage at time k, It is the power relaxation variable of electrochemical energy storage, which allows the energy storage power to increase or decrease by a certain amount on the basis of the original limited range. and are the lower and upper limits of the power of electrochemical energy storage, respectively;

[0132] (2) SOC boundary constraint relaxation:

[0133]

[0134] Where, is the energy storage state of charge at time k, It is a slack variable of the SOC constraint, which is used to allow the energy storage state of charge to have a certain flexibility outside the limited range. and are the lower and upper limits of the energy storage state of charge, respectively;

[0135] (3) Load regulation constraint relaxation:

[0136]

[0137] Where, is the adjustable power of the adjustable load at time k, It is the power slack variable of the adjustable load, which is used to allow the load adjustment power to exceed the original range to a certain extent. and They are the lower and upper limits of the adjustable power of the adjustable load respectively;

[0138] The penalty term is used to limit the slack variable, and is calculated as follows:

[0139]

[0140] Where, is the penalty term, N is the total number of calculation periods, 、 、 They are 、 、 The penalty coefficient is positively correlated with the volatility of new energy. The penalty control of the slack amount is achieved by summing the product of the square of each slack variable and the corresponding penalty coefficient.

[0141] The penalty coefficient is determined as follows:

[0142]

[0143] Where, is the standard deviation of the forecast error of renewable energy output, is the basic penalty coefficient, To adjust the gain factor.

[0144] In a further embodiment of the present invention, the objective function of the hierarchical predictive control model is as follows:

[0145]

[0146]

[0147]

[0148]

[0149] Where, 、 、 and They are the grid stability optimization objective function, the economic optimization objective function, the SOC balance optimization objective function and the SOC variance minimization optimization objective function, respectively. 、 、 and They are 、 、 and The weight of

[0150] The following is an introduction to each optimization objective.

[0151] (1) Grid stability optimization objective function:

[0152]

[0153] Where: is the actual power of the grid at time k, The corrected reference power trajectory is used to improve system stability and other performance by making the actual power of the power grid as close as possible to the corrected reference value.

[0154] (2) Economic optimization objective function:

[0155]

[0156] Where, 、 and They are energy storage operation cost, adjustable load compensation cost and new energy power abandonment cost. 、 and They are 、 and The weight coefficient of .

[0157] Economic optimization objective function It is constructed from three dimensions: energy storage operating cost, adjustable load compensation cost and new energy power abandonment cost.

[0158] The energy storage operating cost includes the charging and discharging loss cost and the equipment life attenuation cost, and the expression can be set as:

[0159]

[0160] Where: is the power loss cost coefficient of the i-th energy storage unit, is the life-cycle cost coefficient, is the SOC change, is the time interval.

[0161] The adjustable load compensation cost is the compensation fee for users to participate in load regulation, and the expression can be:

[0162]

[0163] Where: is the load regulation compensation unit price, It is the adjustable load power variation.

[0164] The cost of curtailed electricity from renewable energy is used to measure the loss of curtailed electricity due to insufficient system regulation capacity. The expression can be set as

[0165]

[0166] Where: is the curtailment cost coefficient, Contribute to new energy, The new energy power absorbed by the system.

[0167] Taking all the above factors into consideration, the proportion of each cost item can be adjusted through the weight coefficient according to actual needs. The weight coefficient can be dynamically adjusted in combination with the average SOC deviation degree of energy storage or the SOC difference between units to achieve coordinated optimization of economic goals and goals such as grid stability and SOC balance.

[0168] (3) SOC balance optimization objective function:

[0169] The SOC balance objective function is used to measure the imbalance degree of the state of charge (SOC) of each energy storage unit in the energy storage system. The specific calculation formula is:

[0170]

[0171] Where M is the total number of energy storage units, is the SOC value of the i-th energy storage unit at time k, is the average SOC value at time k.

[0172] (4) SOC variance minimization optimization objective function:

[0173]

[0174]

[0175] Where, is the real-time average SOC of all energy storage units. The SOC variance minimization term is introduced into the objective function to balance the SOC states among multiple energy storage units.

[0176] In a further embodiment of the present invention, a multi-dimensional state feedback compensation strategy is proposed during the rolling solution of a pre-built hierarchical predictive control model. This strategy combines the feedback of new energy output forecast errors with a dynamic balancing mechanism for energy storage state of charge (SOC) to achieve cross-timescale coordinated control of multiple types of resources. Specifically, the multi-dimensional state feedback compensation strategy forms a closed-loop control loop by collecting multi-dimensional information such as new energy output errors, energy storage SOC status, and load adjustment potential in real time. This strategy dynamically corrects the optimization objectives and constraints of the hierarchical MPC, achieving cross-timescale coordination of multiple types of resources. The core process is as follows:

[0177] 1) Real-time status perception: Obtain the deviation between the actual output of renewable energy and the predicted value, understand the current state of charge (SOC real-time value) of the energy storage, and the status of load adjustment, providing basic data support for subsequent control decisions.

[0178] 2) Use error feedback to correct renewable energy output: Based on the prediction error, the power regulation target value of the subsequent time window is revised in a rolling manner.

[0179] Error quantification formula:

[0180]

[0181] Where: is the actual output of new energy at time k, is the predicted output of new energy at time k.

[0182] And correct the future forecast curve by rolling:

[0183]

[0184] is the error attenuation coefficient, which makes the prediction more consistent with reality and thus adjusts the power regulation target.

[0185] 3) SOC dynamic balancing: Based on the real-time SOC status of each energy storage unit, the power allocation priority is adjusted to balance the SOC status of each energy storage unit, prevent overcharging or over-discharging of some units, and ensure the safe and efficient operation of the energy storage system.

[0186] The energy storage SOC dynamic balancing mechanism generates SOC balancing instructions based on the real-time SOC status of each energy storage unit, and prioritizes the energy storage unit with the largest SOC deviation from the mean.

[0187] The SOC deviation reflects the degree to which the SOC of a single energy storage unit deviates from the overall average level. The specific calculation formula is:

[0188]

[0189] The power allocation weight is calculated based on the SOC deviation of each energy storage unit. The specific calculation formula is:

[0190]

[0191] Where: The numerator is the SOC deviation of the i-th energy storage unit , the denominator is the sum of the SOC deviations of all M energy storage units, Represents the weight of the i-th energy storage unit in power distribution.

[0192] The power instruction correction represents the final power instruction of the i-th energy storage unit at time k. The specific calculation formula is:

[0193]

[0194] In the formula, the basic power And the power correction amount generated by factors such as SOC imbalance, zero error, and safety error 、 、 Together they form.

[0195] SOC dynamic balancing first requires setting a safe SOC operating range. When the SOC of a certain energy storage exceeds this range, a priority reduction strategy is triggered to limit its power allocation ratio to prevent the energy storage unit from degrading due to overcharging / over-discharging. When the limit is exceeded, other healthy energy storage units are preferentially called to ensure continuous operation of the system.

[0196] SOC safe operation range parameter setting:

[0197]

[0198] The over-limit detection monitors the SOC value of each energy storage unit in real time. If one of the following conditions is met, it is determined to be over-limit:

[0199]

[0200] Adopt priority reduction strategy and allocate weight to slightly over-limit power Reduced to 50% of the original value, serious violations will be forced , charging and discharging operations are prohibited, only passive balancing is allowed, and the weight calculation formula is revised:

[0201]

[0202] Where, Assign a weight to the corrected power of the i-th energy storage unit, and are the lower and upper limits of the SOC safe operating range.

[0203] In one embodiment of the present invention, for electrochemical energy storage, flywheel energy storage, and adjustable load, the power allocation priority is as follows:

[0204] The time scale from seconds to minutes is allocated to flywheel energy storage; the time scale from minutes to hours is allocated to electrochemical energy storage; and the time scale above hours is allocated to adjustable loads.

[0205] Specifically, the division of time scales is the key to achieving multi-resource coordination. It is specifically divided into time scales of seconds to minutes, with flywheel energy storage being prioritized to respond to high-frequency power fluctuations; on time scales of minutes to hours, electrochemical energy storage is used to compensate for medium and low-frequency power deviations; on time scales above hours, power supply and demand balance is achieved through the timing flexibility of adjustable loads.

[0206] This invention not only significantly improves the coordinated control accuracy and resource utilization efficiency of multiple types of energy storage and adjustable loads, but also effectively enhances the power grid's ability to smooth out fluctuations in renewable energy and its operational stability. It has dynamic adaptive adjustment capabilities for complex scenarios such as high-frequency power fluctuations and energy storage SOC imbalance, providing strong technical support for flexible and reliable control of high-proportion renewable energy power systems.

[0207] Based on the above embodiments, it can be seen that the present invention shows significant advantages in terms of model construction and coordinated regulation compared to the existing technology. Existing studies often ignore the differences in the characteristics of multiple types of energy storage and adjustable loads. The present invention establishes dynamic characteristic models of electrochemical energy storage, flywheel energy storage, and adjustable loads to accurately quantify and analyze the differences in key parameters such as power regulation rate, capacity boundary, and response lag. For electrochemical energy storage, a second-order RC network equivalent circuit model is used to characterize the polarization effect, a mechanical-electrical coupling equation is constructed for flywheel energy storage, and a piecewise linearization model is designed for adjustable loads to describe power regulation and delayed response characteristics. On this basis, combined with a multi-dimensional state feedback compensation strategy, multi-dimensional information such as new energy output error and energy storage SOC status is collected in real time to form a closed-loop control loop, and the optimization objectives and constraints are dynamically corrected, realizing refined coordinated regulation of multiple types of resources across time scales from seconds to hours. This breaks through the limitations of traditional methods in insufficient handling of resource characteristic differences.

[0208] Secondly, the proposed system is more innovative in its adaptive optimization and robustness enhancement of control strategies. Traditional model predictive control (MPC) suffers from heavy computational burdens and high risk of model mismatch when dealing with high-dimensional nonlinear systems, and its multi-objective optimization weight adjustment lacks dynamics. The improved hierarchical MPC framework constructed in this paper introduces a dynamic weight allocation mechanism and constraint relaxation factors, dynamically adjusting the ratio of stability weight to economic weight based on real-time conditions such as the fluctuation amplitude of renewable energy output and the energy storage SOC level. For example, during periods of significant renewable energy power fluctuation, the stability weight is automatically increased to prioritize grid stability; during periods of stability, the focus is on optimizing economic efficiency and equipment lifespan. Furthermore, slack variables are used to transform hard constraints into flexible constraints, allowing for short-term relaxation when prediction errors are large. A penalty term positively correlated with the renewable energy volatility is used to limit the amount of relaxation, ensuring the solvability of the optimization problem while avoiding excessive deviation from the original constraints. This adaptive adjustment mechanism and robust design significantly enhance the system's adaptability to the uncertainties of renewable energy sources. Compared to traditional methods, it better balances computational efficiency, optimization accuracy, and multi-objective balance, providing a more flexible and reliable control approach for power systems with a high proportion of renewable energy.

[0209] Based on the same inventive concept, the embodiment of the present application also provides an optimized control device for multiple types of energy storage and adjustable loads based on improved model predictive control, which is used to implement the optimized control method for multiple types of energy storage and adjustable loads based on improved model predictive control. The implementation solution provided by the device is similar to the implementation solution described in the above method. Therefore, the specific limitations of the embodiment of the optimized control device for multiple types of energy storage and adjustable loads based on improved model predictive control provided below can be found in the above limitations of the optimized control method for multiple types of energy storage and adjustable loads based on improved model predictive control, and will not be repeated here.

[0210] See also Figure 3 The embodiment of the present invention provides an optimization control device for multiple types of energy storage and adjustable loads based on improved model predictive control, including:

[0211] Model acquisition module, used to obtain dynamic characteristic models of multiple types of energy storage and adjustable loads;

[0212] The model solving module is used to perform rolling solution on the pre-built hierarchical predictive control model based on the parameters provided by the dynamic characteristic model and generate corresponding control instructions;

[0213] An optimization and control module is used to divide control instructions by time scale and output them to the corresponding multi-type energy storage and adjustable loads for execution;

[0214] Among them, the hierarchical predictive control model is used to model the preset optimization objectives and stipulate the power allocation priority. The power allocation priority is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales.

[0215] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0216] Reference Figure 4 An embodiment of the present invention further provides a computer device, comprising: a memory and a processor and a computer program stored in the memory. When the computer program is executed on the processor, it implements the optimization control method of multiple types of energy storage and adjustable loads based on improved model predictive control as described in any one of the above methods.

[0217] The computer device may be a desktop computer, notebook computer, PDA, cloud server or other computing device. The computer device may include, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that Figure 4The computer device is merely an example and does not constitute a limitation on the computer device. The computer device may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, etc.

[0218] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0219] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard drive or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped with the computer device. Furthermore, the memory may include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been output or is about to be output.

[0220] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for optimizing and controlling multiple types of energy storage and adjustable loads based on improved model predictive control as described in any one of the above methods is implemented.

[0221] In this embodiment, if the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a camera / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electric carrier signals or telecommunication signals.

[0222] An embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control as described in any one of the above methods.

[0223] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0224] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0225] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0226] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An optimization control method for multi-type energy storage and adjustable loads based on improved model predictive control, characterized in that: The steps include: Obtain dynamic characteristic models of multiple types of energy storage and adjustable loads; Based on the parameters provided by the dynamic characteristic model, a pre-built hierarchical predictive control model is subjected to rolling solution, and corresponding control instructions are generated; Dividing the control instructions according to time scales and outputting them to the corresponding multiple types of energy storage and adjustable loads for execution; Among them, the hierarchical predictive control model is used to model the preset optimization objectives and stipulate the power allocation priority, and the power allocation priority is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales.

2. The optimization control method for multi-type energy storage and adjustable load based on improved model predictive control according to claim 1 is characterized in that: The objective function of the hierarchical predictive control model is determined by the sum of the products of several optimization objectives and corresponding weights; the weights include at least stability weights and economic weights; the stability weights are used to measure the importance of the optimization objectives of ensuring grid stability in the objective function, and the economic weights are used to reflect the importance of the optimization objectives focusing on economy in the objective function; the stability weights increase with the increase of the rate of change of new energy power, and the economic weights increase with the increase of the degree to which the average SOC of energy storage deviates from the set value and / or the SOC difference between energy storage units.

3. The optimization control method for multi-type energy storage and adjustable loads based on improved model predictive control according to claim 2, characterized in that: If the optimization objectives include grid stability optimization objective, economic optimization objective, SOC balance optimization objective, and SOC variance minimization optimization objective, the corresponding weights are determined according to the following formulas: Where, is the weight of the grid stability optimization objective, belonging to the stability weight, To contribute to new energy, t is time, is a constant; is the weight of the economic optimization objective, belonging to the economic weight, It is a quantitative value of the degree to which the average SOC of the energy storage deviates from the set value; is the weight of the SOC balance optimization target, which belongs to the economic weight. is the quantified value of the maximum SOC difference between energy storage units; The weight of the SOC variance minimization optimization objective belongs to the economic weight; 、 、 and are the preset coefficients of each weight respectively.

4. The optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to claim 2, characterized in that: The constraints of the hierarchical predictive control model include energy storage power constraints, SOC boundary constraints, and load regulation constraints. The energy storage power constraints, SOC boundary constraints, and load regulation constraints are all processed using slack variables, and penalty terms are used to limit each constraint. The constraints and penalty terms after relaxation are as follows: Energy storage power constraint relaxation: Where, is the power of electrochemical energy storage at time k, is the power relaxation variable of electrochemical energy storage, and are the lower and upper limits of the power of electrochemical energy storage, respectively; SOC boundary constraint relaxation: Where, is the energy storage state of charge at time k, is the slack variable of the SOC constraint, and are the lower and upper limits of the energy storage state of charge, respectively; Load regulation constraint relaxation: Where, is the adjustable power of the adjustable load at time k, is the power slack variable of the adjustable load, and They are the lower and upper limits of the adjustable power of the adjustable load respectively; The penalty items: Where, is the penalty term, N is the total number of calculation periods, 、 、 They are 、 、 The penalty coefficient; The penalty coefficient is determined according to the following formula: Where, is the standard deviation of the forecast error of renewable energy output, is the basic penalty coefficient, To adjust the gain factor.

5. The optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to claim 2, characterized in that: The objective function of the hierarchical predictive control model is as follows: Where, 、 、 and They are the grid stability optimization objective function, the economic optimization objective function, the SOC balance optimization objective function and the SOC variance minimization optimization objective function, respectively. 、 、 and They are 、 、 and The weight of N is the total number of calculation periods, is the actual power of the power grid at time k, is the corrected reference power trajectory; 、 and They are energy storage operation cost, adjustable load compensation cost and new energy power abandonment cost. 、 and They are 、 and The weight coefficient of M is the total number of energy storage units, is the SOC value of the i-th energy storage unit at time k, is the average SOC value at time k.

6. The optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to claim 5 is characterized in that: In the rolling solution of the pre-built hierarchical predictive control model, error feedback is used to correct the output of renewable energy. In the SOC balance optimization process, SOC dynamic balancing is used to adjust the power allocation priority of each energy storage unit. The error feedback correction and SOC dynamic balancing are as follows: The error feedback correction is as follows: Where, is the actual output of new energy at time k, is the predicted output of new energy at time k, is the k-time error, is the error attenuation coefficient; Contribute to the revised new energy; The SOC dynamic balance is as follows: The initial power distribution weight of each energy storage unit is calculated as follows: Where, Assign a weight to the initial power of the i-th energy storage unit, is the SOC deviation of the i-th energy storage unit, is the SOC deviation of the j-th energy storage unit; The corrected power distribution weights after SOC dynamic balancing are as follows: Where, Assign a weight to the corrected power of the i-th energy storage unit, and are the lower and upper limits of the SOC safe operating range.

7. The optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to claim 1, characterized in that: For electrochemical energy storage, flywheel energy storage and adjustable loads, the power allocation priorities are as follows: The time scale from seconds to minutes is allocated to the flywheel energy storage; the time scale from minutes to hours is allocated to the electrochemical energy storage; and the time scale above hours is allocated to the adjustable load.

8. An optimized control device for multiple types of energy storage and adjustable loads based on improved model predictive control, characterized in that: include: Model acquisition module, used to obtain dynamic characteristic models of multiple types of energy storage and adjustable loads; A model solving module, configured to perform rolling solving on a pre-built hierarchical predictive control model based on parameters provided by the dynamic characteristic model, and to generate corresponding control instructions; An optimization and control module, configured to divide the control instructions according to time scales and output them to the corresponding multi-type energy storage and adjustable loads for execution; Among them, the hierarchical predictive control model is used to model the preset optimization objectives and stipulate the power allocation priority, and the power allocation priority is used to output the corresponding control instructions to the corresponding energy storage or adjustable load at different time scales.

9. A computer device, characterized in that: The device includes a processor and a memory: The memory is used to store the computer program and send instructions of the computer program to the processor; The processor executes the optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to any one of claims 1 to 7 according to the instructions of the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements an optimization control method for multiple types of energy storage and adjustable loads based on improved model predictive control according to any one of claims 1 to 7.

Images