Comprehensive energy management early warning system and method

Abstract

The invention relates to the technical field of data processing, and discloses a comprehensive energy management early warning system and method.The method comprises the steps that scene recognition and parameter initialization are conducted according to the type of a photovoltaic power station; building a multi-objective optimization model, wherein model decision variables comprise energy storage charging and discharging power, power generation power and electrical load; carrying out dynamic weighting comprehensive optimization on coefficients of the multi-objective optimization model; performing optimization solution, and performing optimization solution on the numerical value comprehensively optimized in the step S3 to obtain a maximum value; outputting the optimal decision variables of the discharge power, the generation power and the electrical load; and the optimal decision variable is dynamically adjusted according to the type of the photovoltaic power station, and early warning information is sent out. According to the invention, through dynamic optimization and scene adaptation rules, abnormal operation is found in time and early warning is carried out, and related operation data is automatically adjusted, so that the power generation efficiency, economy, safety and carbon emission reduction effects of the photovoltaic power station are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an integrated energy management early warning system and method. Background Art

[0002] The current integrated energy and power management system of photovoltaic power stations integrates the development of multiple data fusion directions, forming a comprehensive management system with data-driven as the core and multi-objective optimization as the goal.

[0003] However, the current management system of photovoltaic power plants suffers from a lack of multi-objective collaborative optimization and a balance between power generation, energy storage, and electricity consumption: existing algorithms focus on a single objective, such as maximum power generation, and lack the collaborative optimization of multiple objectives, such as economy, safety, and carbon emission reduction. For example, if the energy storage charging and discharging strategy only pursues peak-valley price arbitrage, it may ignore the loss of equipment life. Insufficient scenario differentiation: Distributed photovoltaic and centralized photovoltaic differ significantly in operation and maintenance modes and fault types, but universal models are difficult to adapt. For example, rooftop photovoltaics need to consider shadow obstruction, while ground-based power stations are more concerned about component aging.

[0004] Therefore, in order to solve the above problems, a comprehensive energy management early warning system, method and system are needed. Summary of the Invention

[0005] The purpose of this invention is to provide a comprehensive energy management early warning system and method. Through dynamic optimization and scenario adaptation rules, it promptly detects operational anomalies and automatically adjusts relevant operating data, significantly improving the power generation efficiency, economy, safety, and carbon emission reduction of photovoltaic power plants.

[0006] The present invention is achieved in that:

[0007] The present invention provides a comprehensive energy management early warning method, which is specifically performed according to the following steps:

[0008] S1: First, according to the type of photovoltaic power station, scene recognition and parameter initialization are performed. In scene recognition and parameter initialization, the photovoltaic power station type S is first input, and distributed or centralized is selected. Then, the target weight vector is constructed as follows:

[0009] ω=[ω1,ω2,ω3,ω4]

[0010] Among them, ω is the weight vector, ω1 is the power generation, ω2 is the economy, ω3 is the safety, and ω4 is the carbon emission reduction; the distributed photovoltaic power station is S=1, and the centralized photovoltaic power station is S=2.

[0011] S2: Construct a multi-objective optimization model, where the model decision variables include energy storage charging and discharging power, power generation power, and power load;

[0012] S2.1: First, establish the objective function of maximizing power generation, as follows:

[0013]

[0014] Among them, P pv (t) is the generated power; f1 is the objective function for maximizing the power generation;

[0015] S2.2: Then establish the objective function for the optimal economic performance of peak-to-valley price difference arbitrage, as follows:

[0016]

[0017] Among them, P ess (t) is the energy storage charging and discharging power, t is the time period, and f2 is the objective function for the optimal economic performance of peak-valley price difference arbitrage;

[0018] S2.3: Then establish the objective function of minimizing equipment life loss and maximizing safety, as shown below;

[0019]

[0020] Where, α·DOD(t) 2 is the energy storage depth discharge, ΔT(t) 2 is the component temperature fluctuation data, α and β are loss coefficients, and f3 is the objective function for minimizing equipment life loss;

[0021] S2.4: Then establish the objective function for maximizing carbon emission reduction;

[0022]

[0023] Among them, η grid Δt is the carbon emission factor of the power grid;

[0024] S2.5: Establish constraints, including energy storage power limit, energy storage SOC balance, and power generation-consumption balance. The energy storage power limit is as follows:

[0025] P ess,min ≤P ess (t)≤P ess,max

[0026] The energy storage SOC balance is as follows:

[0027] SOC min ≤SOC(t)≤SOC max

[0028] The power generation-consumption balance is as follows:

[0029] P pv (t)+Pess (t) = P load (t);

[0030] S3: Dynamically weighted comprehensive optimization of the coefficients of the multi-objective optimization model;

[0031] The coefficients of the multi-objective optimization model are dynamically weighted and comprehensively optimized, as shown in the following formula:

[0032]

[0033] Among them, f i,max is the normalization coefficient of a single target, f1,i, f2,i, f3,i, f4,i are the power generation of particle i, and f2 is the target value of economy, safety and carbon emission reduction.

[0034] S4: Optimize and solve the value comprehensively optimized in step S3 to find the maximum value; specifically, perform the following steps:

[0035] First, update the historical optimal position p of each particle best,i , and the global optimal position g best ;

[0036] like Then p best,i =x i ;

[0037] like Then g best =x i ;

[0038] Then update the particle speed and position according to the individual and global optimality. The speed update is as follows:

[0039]

[0040] The position update is as follows:

[0041]

[0042] Among them, ω is the weight vector, c1 and c2 are learning factors; r1 and r2 are random numbers;

[0043] S5: optimal decision variables for output discharge power, generated power, and power load;

[0044] When the maximum number of iterations K is reached max , terminate the iteration and output the global optimal position g best The corresponding decision variables are as follows:

[0045]

[0046] S6: Then dynamically adjust the optimal decision variables according to the type of photovoltaic power station and issue early warning information. If distributed photovoltaic S=1, add a shadow occlusion detection module and dynamically adjust P pv (t), introduce the component health threshold, when the health <H threshold When forced to reduce P pv (t);

[0047] Centralized photovoltaic S=2, add component aging prediction model, dynamically adjust ω1 and ω3, introduce energy storage life prediction model, when energy storage life <L threshold When limiting P ess (t) Fluctuation range.

[0048] Furthermore, the present invention provides a comprehensive energy management early warning system, including a scenario recognition and parameter configuration module. This system automatically identifies the current operating scenario through light intensity, temperature, wind speed sensor data, distributed / centralized power station type, external data electricity price policy, and weather forecasts. It also dynamically adjusts optimization target weights based on the scenario, including power generation, economy, safety, carbon emission reduction, and constraints such as energy storage SOC range and equipment life threshold.

[0049] The rule engine and strategy library module dynamically adjusts the optimization strategy based on preset rules;

[0050] Multi-objective optimization model building module: Dynamically builds a multi-objective optimization model that includes power generation, economy, safety, and carbon emission reduction based on scenarios and rule engine output;

[0051] Optimization solver module, runs the optimization algorithm, and solves the comprehensive objective function F total The maximum value of

[0052] Decision execution and monitoring module: Converts optimal decision variables into control instructions, including energy storage charging and discharging power, power generation power adjustment, real-time monitoring, tracking decision execution effects including power generation and economic indicators, and recording deviations between actual and predicted values; then performs exception processing, and triggers an alarm when detecting execution deviations and if the energy storage fails to respond to instructions.

[0053] Furthermore, the present invention provides a computer-storable medium, wherein the storage medium stores a computer program. When the computer program in the storage medium runs, any one of the above-mentioned comprehensive energy management early warning methods is executed.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. This invention monitors the energy storage system's charge and discharge power, power generation, and load, and dynamically adjusts weights to prevent overcharging and over-discharging. For example, when the storage SOC falls below 20%, the algorithm automatically limits discharge power, extending battery life by over 30%. Furthermore, by charging during low-price periods (such as nighttime) and discharging during high-price periods (such as daytime), combined with demand response mechanisms, peak-valley arbitrage profits can be increased by 20%-40%.

[0056] Rapid fault response: Combined with real-time data streams (such as current and voltage anomalies, temperature changes), the present invention can trigger fault warnings within 5 seconds and automatically execute data adjustment protection strategies to reduce accident losses.

[0057] 2. The present invention monitors various data of photovoltaic power generation in real time according to different photovoltaic power generation sites, issues timely warnings when abnormalities are found, and dynamically adjusts the energy storage charging and discharging strategy according to the model based on the abnormal data adjusted by the environment.

[0058] Reduced operation and maintenance costs: Through predictive maintenance (such as component aging prediction and inverter failure warning), the algorithm can reduce unplanned downtime and reduce operation and maintenance costs by 10%-20%. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It is understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 is a flow chart of the method of the present invention;

[0061] Figure 2 It is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, 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 described embodiments are 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. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention for which protection is sought, but is merely for selected embodiments of the present invention. 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.

[0063] See also Figure 1-Figure 2 The present invention provides a comprehensive energy management early warning method, which is specifically performed according to the following steps:

[0064] S1: First, according to the type of photovoltaic power station, scene recognition and parameter initialization are performed. In scene recognition and parameter initialization, the photovoltaic power station type S is first input, and distributed or centralized is selected. Then, the target weight vector is constructed as follows:

[0065] ω=[ω1,ω2,ω3,ω4]

[0066] Among them, ω is the weight vector, ω1 is the power generation, ω2 is the economy, ω3 is the safety, and ω4 is the carbon emission reduction; the distributed photovoltaic power station is S=1, and the centralized photovoltaic power station is S=2.

[0067] S2: Construct a multi-objective optimization model, where the model decision variables include energy storage charging and discharging power, power generation power, and power load;

[0068] S2.1: First, establish the objective function of maximizing power generation, as follows:

[0069]

[0070] Among them, P pv (t) is the generated power; f1 is the objective function for maximizing the power generation;

[0071] S2.2: Then establish the objective function for the optimal economic performance of peak-to-valley price difference arbitrage, as follows:

[0072]

[0073] Among them, P ess (t) is the energy storage charging and discharging power, t is the time period, and f2 is the objective function for the optimal economic performance of peak-valley price difference arbitrage;

[0074] S2.3: Then establish the objective function of minimizing equipment life loss and maximizing safety, as shown below;

[0075]

[0076] Where, α·DOD(t) 2 is the energy storage depth discharge, ΔT(t) 2 is the component temperature fluctuation data, α and β are loss coefficients, and f3 is the objective function for minimizing equipment life loss;

[0077] S2.4: Then establish the objective function for maximizing carbon emission reduction;

[0078]

[0079] Among them, η grid Δt is the carbon emission factor of the power grid;

[0080] S2.5: Establish constraints, including energy storage power limit, energy storage SOC balance, and power generation-consumption balance. The energy storage power limit is as follows:

[0081] P ess,min ≤P ess (t)≤P ess,max

[0082] The energy storage SOC balance is as follows:

[0083] SOC min ≤SOC(t)≤SOC max

[0084] The power generation-consumption balance is as follows:

[0085] P pv (t)+P ess (t) = P load (t);

[0086] S3: Dynamically weighted comprehensive optimization of the coefficients of the multi-objective optimization model;

[0087] The coefficients of the multi-objective optimization model are dynamically weighted and comprehensively optimized, as shown in the following formula:

[0088]

[0089] Among them, f i,max is the normalization coefficient of a single target, f1,i, f2,i, f3,i, f4,i are the power generation of particle i, and f2 is the target value of economy, safety and carbon emission reduction.

[0090] S4: Optimize and solve the value comprehensively optimized in step S3 to find the maximum value; specifically, perform the following steps:

[0091] First, update the historical optimal position p of each particle best,i , and the global optimal position g best ;

[0092] like Then p best,i =x i ;

[0093] like Then g best =x i ;

[0094] Then update the particle speed and position according to the individual and global optimality. The speed update is as follows:

[0095]

[0096] The position update is as follows:

[0097]

[0098] Among them, ω is the weight vector, c1 and c2 are learning factors; r1 and r2 are random numbers;

[0099] S5: optimal decision variables for output discharge power, generated power, and power load;

[0100] When the maximum number of iterations K is reached max , terminate the iteration and output the global optimal position g best The corresponding decision variables are as follows:

[0101]

[0102] S6: Then dynamically adjust the optimal decision variables according to the type of photovoltaic power station and issue early warning information. If distributed photovoltaic S=1, add a shadow occlusion detection module and dynamically adjust P pv (t), introduce the component health threshold, when the health <H threshold When forced to reduce P pv (t);

[0103] Centralized photovoltaic S=2, add component aging prediction model, dynamically adjust ω1 and ω3, introduce energy storage life prediction model, when energy storage life <L threshold When limiting P ess (t) Fluctuation range.

[0104] In this embodiment, the present invention provides a comprehensive energy management early warning system, including a scenario recognition and parameter configuration module; automatically identifying the current operating scenario through light intensity, temperature, wind speed sensor data, distributed / centralized power station type, external data electricity price policy, and weather forecast; and dynamically adjusting optimization target weights based on the scenario, including power generation, economy, safety, carbon emission reduction, and constraints such as energy storage SOC range and equipment life threshold;

[0105] The rule engine and strategy library module dynamically adjusts the optimization strategy based on preset rules;

[0106] Multi-objective optimization model building module: Dynamically builds a multi-objective optimization model that includes power generation, economy, safety, and carbon emission reduction based on scenarios and rule engine output;

[0107] Optimization solver module, runs the optimization algorithm, and solves the comprehensive objective function F total The maximum value of

[0108] Decision execution and monitoring module: Converts optimal decision variables into control instructions, including energy storage charging and discharging power, power generation power adjustment, real-time monitoring, tracking decision execution effects including power generation and economic indicators, and recording deviations between actual and predicted values; then performs exception processing, and triggers an alarm when detecting execution deviations and if the energy storage fails to respond to instructions.

[0109] In this embodiment, the present invention provides a computer-storable medium, wherein the storage medium stores a computer program. When the computer program in the storage medium runs, any one of the above-mentioned comprehensive energy management early warning methods is executed.

[0110] In this embodiment, taking a rooftop distributed photovoltaic power station as an example, the meteorological and power generation data are shown in Tables 1 to 6;

[0111] Table 1 Meteorological and power generation data

[0112]

[0113] Table 2 Electricity price and load data

[0114] Time period t Electricity price C(t) (yuan / kWh) <![CDATA[Electrical load P load,demand (t) (kW)]]> 1 0.3 80 8 0.8 150 12 1.2 200 18 0.6 100

[0115] Table 3 Energy storage system data

[0116] parameter Numerical Energy storage capacity Eess (kWh) 200 Initial SOC 50％ <![CDATA[Maximum charge and discharge power P ess , max (kW)]]> 50

[0117] Table 4 Optimized decision variables

[0118]

[0119] Table 6 Key indicator data

[0120]

[0121]

[0122] In this embodiment, the multi-objective coordination of this embodiment improves the economic efficiency (increase in peak-valley arbitrage income) and carbon emission reduction effect while ensuring power generation through dynamic weight adjustment, and at the same time reduces the energy storage life loss.

[0123] Scene adaptation: During low-light periods (such as t=1), the present invention increases the weight of carbon emission reduction and gives priority to the use of clean energy; during low SOC periods of energy storage (such as t=18), the algorithm limits discharge power to protect battery life.

[0124] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A comprehensive energy management early warning method, characterized by: Follow these steps: S1: First, perform scene recognition and parameter initialization according to the type of photovoltaic power station; S2: Construct a multi-objective optimization model, where the model decision variables include energy storage charging and discharging power, power generation power, and power load; S3: Dynamically weighted comprehensive optimization of the coefficients of the multi-objective optimization model; S4: Optimize and solve the value comprehensively optimized in step S3 to find the maximum value; S5: optimal decision variables for output discharge power, generated power, and power load; S6: Then dynamically adjust the optimal decision variables according to the type of photovoltaic power station and issue early warning information.

2. The comprehensive energy management early warning method according to claim 1, characterized in that: In step S1, during scene recognition and parameter initialization, the photovoltaic power station type S is first input, and distributed or centralized is selected; then the target weight vector is constructed as follows: ω=[ω1,ω2,ω3,ω4] Among them, ω is the weight vector, ω1 is the power generation, ω2 is the economy, ω3 is the safety, and ω4 is the carbon emission reduction; the distributed photovoltaic power station is S=1, and the centralized photovoltaic power station is S=2.

3. The comprehensive energy management early warning method according to claim 1, characterized in that: In step S2, the multi-objective optimization model is constructed by following the steps below: S2.1: First, establish the objective function of maximizing power generation, as follows: Among them, P pv (t) is the generated power; f1 is the objective function for maximizing the power generation; S2.2: Then establish the objective function for the optimal economic performance of peak-to-valley price difference arbitrage, as follows: Among them, P ess (t) is the energy storage charging and discharging power, t is the time period, and f2 is the objective function for the optimal economic performance of peak-valley price difference arbitrage; S2.3: Then establish the objective function of minimizing equipment life loss and maximizing safety, as shown below; Where, α·DOD(t) 2 is the energy storage depth discharge, ΔT(t) 2 is the component temperature fluctuation data, α and β are loss coefficients, and f3 is the objective function for minimizing equipment life loss; S2.4: Then establish the objective function for maximizing carbon emission reduction; Among them, η grid Δt is the carbon emission factor of the power grid; S2.5: Establish constraints, including energy storage power limit, energy storage SOC balance, and power generation-consumption balance. The energy storage power limit is as follows: P ess,min ≤P ess (t)≤P ess,max The energy storage SOC balance is as follows: SOC min ≤SOC(t)≤SOC max The power generation-consumption balance is as follows: P pv (t)+P ess (t)=P load (t)。 4. The comprehensive energy management early warning method according to claim 1, characterized in that: In step S3, the coefficients of the multi-objective optimization model are dynamically weighted and comprehensively optimized, as shown in the following formula: Among them, f i,max is the normalization coefficient of a single target, f1,i, f2,i, f3,i, f4,i are the power generation of particle i, and f2 is the target value of economy, safety and carbon emission reduction.

5. The comprehensive energy management early warning method according to claim 1, characterized in that: In step S4, the following steps are specifically performed: First, update the historical optimal position p of each particle best,i , and the global optimal position g best ; like Then p best,i =x i ; like Then g best =x i ; Then update the particle speed and position according to the individual and global optimality. The speed update is as follows: The position update is as follows: Among them, ω is the weight vector, c1 and c2 are learning factors; r1 and r2 are random numbers; When the maximum number of iterations K is reached max , terminate the iteration and output the global optimal position g best The corresponding decision variables are as follows:

6. The comprehensive energy management early warning method according to claim 1, characterized in that: In step S6, if distributed photovoltaic S=1, a shadow detection module is added to dynamically adjust P pv (t), introduce the component health threshold, when the health <H threshold When forced to reduce P pv (t); Centralized photovoltaic S=2, add component aging prediction model, dynamically adjust ω1 and ω3, introduce energy storage life prediction model, when energy storage life <L threshold When limiting P ess (t) Fluctuation range.

7. An integrated energy management early warning system, characterized by: It includes a scenario recognition and parameter configuration module. It automatically identifies the current operating scenario through light intensity, temperature, wind speed sensor data, distributed / centralized power station type, external data electricity price policy, and weather forecast. It also dynamically adjusts the optimization target weights based on the scenario, including power generation, economy, safety, carbon emission reduction, and constraints such as energy storage SOC range and equipment life threshold. The rule engine and strategy library module dynamically adjusts the optimization strategy based on preset rules; Multi-objective optimization model building module: Dynamically builds a multi-objective optimization model that includes power generation, economy, safety, and carbon emission reduction based on scenarios and rule engine output; Optimization solver module, runs the optimization algorithm, and solves the comprehensive objective function F total The maximum value of Decision execution and monitoring module: Converts optimal decision variables into control instructions, including energy storage charging and discharging power, power generation power adjustment, real-time monitoring, tracking decision execution effects including power generation and economic indicators, and recording deviations between actual and predicted values; then performs exception processing, and triggers an alarm when detecting execution deviations and if the energy storage fails to respond to instructions.

8. A computer storable medium, characterized in that: The storage medium stores a computer program. When the computer program in the storage medium is run, the comprehensive energy management early warning method described in any one of claims 1 to 6 is executed.

Images