Energy management system and multi-source collaborative optimization scheduling method for multi-energy complementary microgrid

Abstract

The present application relates to the technical field of smart grid and energy management, in particular to a multi-energy complementary microgrid energy management system and a multi-source collaborative optimization scheduling method. It comprises: an energy monitoring unit for real-time monitoring of equipment operation and energy consumption in the microgrid; an energy management unit based on differential geometry manifold theory, optimizes the allocation of each energy by constructing a multi-dimensional manifold of energy flow; a load prediction unit for monitoring equipment load conditions, reducing the load of faulty equipment according to the fault prediction index; an abnormality detection unit based on the impact response matrix to detect abnormal events, and the abnormal events are fed back in stages. The multi-energy complementary microgrid energy management system and multi-source collaborative optimization scheduling method combine probability control optimization technology through differential geometry manifold theory and adaptive weight mapping, realize real-time optimization and dynamic distribution of multi-energy output, and can quickly respond under abnormal events and equipment aging conditions, improving the energy utilization efficiency of the microgrid.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart grid and energy management, in particular, to an energy management system of a multi-energy complementary microgrid and a multi-source collaborative optimization scheduling method. BACKGROUND

[0002] The energy management system of the multi-energy complementary microgrid and the multi-source collaborative optimization scheduling method aim to realize efficient energy distribution and reliable and stable operation of the system, control the output proportion of each energy device and load distribution by combining the dynamic distribution model based on differential geometric manifold and the adaptive weight mapping with the probability control optimization technology, and realize dynamic optimization scheduling according to the aging state of the device, abnormal event response and real-time load change.

[0003] The existing microgrid energy management system and optimization scheduling method are usually difficult to maintain stable and efficient energy distribution in a renewable energy environment with high volatility, and due to the aging of microgrid devices and the complex situation of load surge, the load of aging devices is too heavy, the energy utilization efficiency is low, and the system is slow to respond in emergency situations. Therefore, the energy management system of the multi-energy complementary microgrid and the multi-source collaborative optimization scheduling method are provided. SUMMARY

[0004] The present application aims to provide an energy management system of a multi-energy complementary microgrid and a multi-source collaborative optimization scheduling method to solve the problems of excessive load of aging devices, low energy utilization efficiency and slow response of the system in emergency situations due to the aging of microgrid devices and the complex situation of load surge as described in the background.

[0005] To achieve the above-mentioned purpose, the present application aims to provide an energy management system of a multi-energy complementary microgrid, comprising:

[0006] An energy monitoring unit for real-time monitoring of device operation and energy consumption in the microgrid;

[0007] Further comprising an energy management unit, which optimizes the distribution of each energy by constructing a multi-dimensional manifold of energy flow based on the theory of differential geometric manifold;

[0008] Further comprising a load prediction unit for monitoring device load conditions, defining and calculating a fault prediction index, predicting device failure based on the fault prediction index and dynamically reducing the load of the failed device;

[0009] Further comprising an anomaly detection unit based on an impact response matrix to detect abnormal events and provide hierarchical feedback of abnormal events.

[0010] The energy monitoring unit comprises a data acquisition module and an energy monitoring module;

[0011] The data acquisition module uses sensors and communication technology to acquire the equipment operation data and the equipment operation state in the micro-grid in real time.

[0012] The equipment operation data includes equipment load, equipment operation duration, equipment current, equipment voltage and equipment temperature.

[0013] The energy monitoring module is used for receiving and displaying the equipment operation data and the equipment operation state in real time.

[0014] As a further improvement of the technical solution, the energy management unit comprises an energy multi-dimensional manifold construction module and an energy optimal distribution module.

[0015] The energy multi-dimensional manifold construction module is based on the differential geometry manifold theory to construct an energy multi-dimensional manifold , generate a manifold feature vector , and the specific method steps are as follows:

[0016] S2.1.1, define the solar power output as , the wind power output as , the energy storage remaining capacity as , and the gas power output as ;

[0017] Define the solar energy utilization efficiency as , the wind energy utilization efficiency as , the energy storage utilization efficiency as , and the gas power utilization efficiency as ;

[0018] Define the total load demand as , is the solar energy distribution weight, is the wind energy distribution weight, is the energy storage distribution weight, is the gas power distribution weight, and satisfies ;

[0019] S2.1.2, construct the state point at time :

[0020] ;

[0021] wherein, is the solar power output at time , the wind energy utilization efficiency at time , the energy storage remaining capacity at time , and the gas power output at time ​​​The gas power generation output at the moment;

[0022] S2.1.3, all moment state points change with time The set of energy multidimensional manifold :

[0023] ;

[0024] Wherein, is the observation period; is the starting time of the period; is the end time of the period;

[0025] S2.1.4, calculate The total output at the moment And manifold feature vector :

[0026] ;

[0027] ;

[0028] Wherein, is the proportion of solar energy in the total output; is the proportion of wind energy in the total output; is the proportion of energy storage in the total output; is the proportion of gas power generation in the total output.

[0029] As a further improvement of the technical solution, the energy optimization distribution module establishes an optimization system total loss objective function based on the manifold feature vector , uses Lagrange multiplier method to solve the energy distribution weight, and distributes energy based on the energy distribution weight. The specific method steps are as follows:

[0030] S2.2.1, based on the total load demand And the utilization efficiency of each energy, the optimization system total loss objective function is established:

[0031] ;

[0032] Wherein, is the total load demand At the moment;

[0033] S2.2.2, construct Lagrange function , use Lagrange multiplier method to solve the distribution weight of each energy:

[0034] ;

[0035] S2.2.3, get the optimal distribution weight of solar energy , wind energy optimal allocation weight , energy storage optimal allocation weight , and gas power generation optimal allocation weight , determine the output proportion of each energy.

[0036] As a further improvement of the technical solution, the load prediction unit comprises a fault detection module and a load adjustment module;

[0037] Wherein, the fault detection module monitors the load condition of the equipment to calculate the fault prediction index And judge whether the equipment has a fault risk, the specific method steps are as follows:

[0038] S3.1.1, according to the current load of the equipment And the maximum load , define and calculate the fault prediction index :

[0039] ;

[0040] S3.1.2, set the fault risk threshold , compare the fault prediction index With the fault risk threshold :

[0041] If , trigger load adjustment and generate load adjustment signal;

[0042] If , no need to adjust the load;

[0043] The load adjustment module receives the load adjustment signal of the fault detection module, dynamically adjusts the load, and the specific method steps are as follows:

[0044] S3.2.1, according to the load adjustment coefficient Calculate the adjusted load :

[0045] ;

[0046] Wherein, The load adjustment coefficient is ;

[0047] S3.2.2, apply the adjusted load To the equipment.

[0048] As a further improvement of the technical solution, the abnormality detection unit comprises an event identification module and a hierarchical feedback module;

[0049] The event identification module identifies and analyzes the abnormal event of the device by constructing an impact response matrix, and generates event type information. The specific steps are as follows:

[0050] S4.1.1, define the device running state value of the first device in the microgrid at the moment as , including device running data, i.e. device load, device running time, device current, device voltage and device temperature; The normal state threshold of the first device is ; the deviation degree of the first device at the moment is

[0051] S4.1.2, based on the device running state value of the first device at the moment and the normal state threshold of the first device , the deviation degree of the first device at the moment is calculated:

[0052]

[0053] ;

[0054] S4.1.3, define an abnormal threshold , if the deviation degree of the first device at the moment is greater than the abnormal threshold , then for the first device, construct an impact response matrix : ;

[0055] ;

[0056] Wherein, is an element in the impact response matrix , indicating the influence degree of the first device in the abnormal state on the second device; is a proportional coefficient; is the influence weight of the first device in the abnormal state on the second device, ; S4.1.4, fuse the impact response matrix of all devices

[0057] ​​​​​​​​​​​​​​​​​, construct the total impact vector :

[0058] .

[0059] As a further improvement of the technical solution, the hierarchical feedback module is based on the total impact vector , the abnormal event is fed back, and the specific steps are as follows:

[0060] S4.2.1, define two event level thresholds and ;

[0061] S4.2.2, according to the total impact vector , the event is divided into three levels:

[0062] Low-level event: ;

[0063] Medium-level event: ;

[0064] High-level event: ;

[0065] S4.2.3, generate a feedback signal according to the event level , and feed back the feedback signal to the energy management system;

[0066] ;

[0067] Among them, represents a low-level event; represents a medium-level event; represents a high-level event.

[0068] On the other hand, the present application provides a multi-source collaborative optimization scheduling method of an energy management system of a multi-energy complementary microgrid, which is used in the energy management system of the multi-energy complementary microgrid, and includes the following steps:

[0069] S8.1, according to the cumulative use time of the device and the current load condition, an exponential decay model is used to calculate the device aging factor;

[0070] In the S8.1, the utilization efficiency, the basic allocation weight and the load demand of each energy device in the energy management system of the multi-energy complementary microgrid are obtained, and according to the cumulative use time of the device and the current load condition, an exponential decay model is used to calculate the device aging factor, and the specific method steps are as follows:

[0071] S8.1.1, obtain the utilization efficiency of each energy , the basic allocation weight of each energy , load demand , each energy output and feedback signal ;

[0072] S8.1.2, according to the cumulative use of the device and the current load conditions, the application of exponential decay model calculation device aging factor :

[0073] ;

[0074] wherein, is the number of energy equipment variable, , when take solar device aging factor, and so on; is the aging rate coefficient of energy equipment; is the aging rate adjustment function; is the integral variable;

[0075] S8.2, combined with the device aging factor to build energy scheduling differential equation, solve the energy scheduling differential equation, get energy correction weight;

[0076] In the S8.2, combined with the device aging factor to build energy scheduling differential equation, solve the energy scheduling differential equation, get energy correction weight, the specific steps are as follows:

[0077] S8.2.1, combined with the device aging factor build energy scheduling differential equation:

[0078] ;

[0079] wherein, is the aging influence adjustment coefficient of energy equipment; is the energy correction weight;

[0080] S8.2.2, solve the energy scheduling differential equation to get energy correction weight .

[0081] S8.3, based on the energy correction weight and feedback signal, introduce adaptive weight mapping function to define the distribution strategy of energy output, through multiple probability control technology, establish distribution optimization probability control objective function, and use Lagrange multiplier method to solve and obtain the optimal distribution weight;

[0082] In the S8.3, based on the energy correction weight and feedback signal An adaptive weight mapping function is introduced to define an energy output allocation strategy. Through multiple probabilistic control techniques, an allocation optimization probabilistic control objective function is established, and the optimal allocation weights are obtained using the Lagrange multiplier method. The specific method is as follows:

[0083] S8.3.1, Based on the feedback signal Construct a weighted smoothing nonlinear function :

[0084] ;

[0085] in, For event response sensitivity;

[0086] S8.3.2, Energy-based weighting Construct an adaptive weight mapping function for each energy source:

[0087] ;

[0088] in, for Optimal weight allocation at any given time; This is the adjustment coefficient;

[0089] S8.3.3, based on Optimal weight allocation at any given time Energy utilization efficiency Weighting of each energy base Load demand Energy output and feedback signals Establish the allocation optimization probabilistic control objective function :

[0090] ;

[0091] in, To allocate and optimize the probabilistic control objective function; For adjustment terms;

[0092] S8.3.4 Constructing the Lagrange function The Lagrange multiplier method is used to solve the objective function of minimizing loss, thus obtaining the optimal weight allocation. :

[0093] ;

[0094] in, It is a Lagrange multiplier.

[0095] S8.4, the optimal allocation weight is applied to the energy management system of the multi-energy complementary microgrid to dynamically allocate the output of each energy source. Compared with the prior art, the present application has the following advantages:

[0096] 1. In the energy management system of the multi-energy complementary microgrid and the multi-source collaborative optimization scheduling method, the energy multi-dimensional manifold model constructed based on the differential geometric manifold theory dynamically reflects the real-time state of the output of each energy source, optimizes the allocation weight of each energy source in the total load, and enables the system to adapt to the changes of fluctuating renewable energy, thereby maintaining efficient and stable energy allocation under different load conditions.

[0097] 2. In the energy management system of the multi-energy complementary microgrid and the multi-source collaborative optimization scheduling method, the adaptive weight mapping combined with the probability control optimization technology is used to realize real-time response and hierarchical scheduling of equipment aging and abnormal events, so that the system can timely adjust the energy allocation strategy under the conditions of weakened output of aging equipment or sudden surge of load. BRIEF DESCRIPTION OF DRAWINGS

[0098] Figure 1 is the overall flowchart of the present application;

[0099] The meanings of the various numbers in the figure are as follows:

[0100] 1. Energy monitoring unit; 2. Energy management unit; 3. Load prediction unit; 4. Abnormality detection unit. DETAILED DESCRIPTION

[0101] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0102] Embodiment 1

[0103] Please refer to Figure 1 as shown, an energy management system of a multi-energy complementary microgrid is provided, which includes:

[0104] The energy monitoring unit 1 is used to monitor the operation of the equipment in the microgrid and the energy consumption in real time. The energy monitoring unit 1 includes a data acquisition module and an energy monitoring module.

[0105] The data acquisition module uses sensors and communication technology to acquire the equipment operation data and the equipment operation state in the microgrid in real time.

[0106] The equipment operation data includes equipment load, equipment operation time, equipment current, equipment voltage, and equipment temperature.

[0107] The device includes a solar energy device, a wind energy device, an energy storage device, and a gas power generation device.

[0108] The energy monitoring module is configured to receive and display the device operation data and the device operation state in real time.

[0109] The energy management unit 2 is configured to optimize the distribution of each energy based on the theory of differential geometric manifold, by constructing a multi-dimensional manifold of energy flow, and the energy management unit 2 includes an energy multi-dimensional manifold construction module and an energy optimized distribution module.

[0110] The energy multi-dimensional manifold construction module is configured to construct an energy multi-dimensional manifold based on the theory of differential geometric manifold. The energy multi-dimensional manifold construction module is configured to generate a manifold feature vector. The specific method steps are as follows:

[0111] S2.1.1, define the solar energy output as , the wind energy output as , the energy storage remaining capacity as , and the gas power generation output as .

[0112] Define the solar energy utilization efficiency as , the wind energy utilization efficiency as , the energy storage utilization efficiency as , and the gas power generation utilization efficiency as .

[0113] Define the total load demand as , , the solar energy distribution weight as , the wind energy distribution weight as , the energy storage distribution weight as , the gas power generation distribution weight as , and satisfy .

[0114] S2.1.2, construct the state point at time :

[0115] .

[0116] Wherein, is the solar energy output at time , the wind energy utilization efficiency at time , the energy storage remaining capacity at time , and the gas power generation output at time .

[0117] ​​S2.1.3, all-time state points over time constitute the energy multidimensional manifold :

[0118]

[0119] wherein, is the observed time period; is the time period start time; is the time period end time;

[0120] S2.1.4, calculating total output at time point and manifold feature vector :

[0121]

[0122]

[0123] wherein, is the proportion of solar energy in the total output; is the proportion of wind energy in the total output; is the proportion of energy storage in the total output; is the proportion of gas power generation in the total output.

[0124] In the present embodiment, the energy multidimensional manifold is a geometric structure representing the state of multiple energy outputs in a high-dimensional space, and the energy multidimensional manifold is constructed by state points p of each energy, each state point p representing the energy state of the system at a certain time point; the energy multidimensional manifold dynamically reflects the proportion of different energy outputs in the system and the change trend thereof.

[0125] The energy optimization distribution module establishes an optimization system total loss objective function based on the manifold feature vector , solves the energy distribution weight using the Lagrange multiplier method, and distributes energy based on the energy distribution weight, and the specific method steps are as follows:

[0126] S2.2.1, based on the total load demand and the utilization efficiency of each energy, an optimization system total loss objective function is established:

[0127]

[0128] wherein, is the total load demand at time point

[0129] ​​​​​S2.2.2, constructing a Lagrange function , and solving the distribution weight of each energy source by using the Lagrange multiplier method:

[0130] ;

[0131] S2.2.3, obtaining the optimal distribution weight of solar energy , the optimal distribution weight of wind energy , the optimal distribution weight of energy storage , and the optimal distribution weight of gas power generation , to determine the output proportion of each energy source.

[0132] In this embodiment, the manifold feature vector is a simplified representation of the energy multi-dimensional manifold , which is used to summarize the proportion of each energy output in the system and serves as the input for subsequent optimization calculation;

[0133] The energy multi-dimensional manifold is constructed by the state points of each energy source at each time point , and can be updated in real time as the time and actual situation change. The manifold feature vector can instantaneously reflect the relative proportion of each energy output in the system, dynamically allocate the output of each energy source, and is suitable for renewable energy with fluctuating characteristics;

[0134] Traditional methods are usually based on fixed energy distribution rules or static weights, which are difficult to effectively adapt to scenarios with large fluctuations in energy output, resulting in low energy utilization efficiency or load imbalance.

[0135] The load prediction unit 3 is used to monitor the equipment load, define and calculate the fault prediction index, predict the equipment failure based on the fault prediction index, and dynamically reduce the load of the faulty equipment. The load prediction unit 3 includes a fault detection module and a load adjustment module.

[0136] In this embodiment, the fault detection module monitors the load of the equipment to calculate the fault prediction index , and judges whether the equipment has a risk of failure. The specific method steps are as follows:

[0137] S3.1.1, according to the current load and the maximum bearing load of the equipment, define and calculate the fault prediction index :

[0138] ;

[0139] S3.1.2, set the fault risk threshold , and compare the fault prediction index with the fault risk threshold value comparison is made:

[0140] if , a load adjustment is triggered, and a load adjustment signal is generated;

[0141] if , the load does not need to be adjusted;

[0142] The load adjustment module receives the load adjustment signal of the fault detection module, and dynamically adjusts the load, and the specific method steps are as follows:

[0143] S3.2.1, according to the load adjustment coefficient Calculate the adjusted load :

[0144] ;

[0145] Wherein, is the load adjustment coefficient, ;

[0146] S3.2.2, apply the adjusted load to the device.

[0147] In this embodiment, the load adjustment coefficient is set according to the characteristics of the device.

[0148] The anomaly detection unit 4 detects abnormal events based on the impact response matrix, and feeds back the abnormal events in stages, and the anomaly detection unit 4 includes an event identification module and a staged feedback module;

[0149] In this embodiment, the event identification module identifies and analyzes the abnormal events of the device by constructing the impact response matrix, and generates event type information, and the specific steps are as follows:

[0150] S4.1.1, define the device running state value of the first device in the microgrid at the time as , including device running data, that is, device load, device running time, device current, device voltage and device temperature;

[0151] The normal state threshold value of the first device is ; the deviation degree of the first device at the time ;

[0152] S4.1.2, based on the first device at the time ​The device operating status value at any given time is and the The normal state threshold for each device is: Calculate the first The device is deviation in time :

[0153] ;

[0154] S4.1.3, Define the anomaly threshold If the first The device is deviation in time Greater than the abnormal threshold Then for the first Each device constructs an impact response matrix. :

[0155] ;

[0156] in, Impact response matrix The element in represents the first element. The device in abnormal condition affects the first The extent of the impact on each device; This is the proportionality coefficient; For the first The device in abnormal condition affects the first The influence weight of each device ;

[0157] S4.1.4, Integrate the impact response matrix of all devices. Construct the total impact vector :

[0158] .

[0159] The graded feedback module is based on the total impact vector. The abnormal events will be categorized and reported in a tiered manner. The specific steps are as follows:

[0160] S4.2.1 Define two event level thresholds and ;

[0161] S4.2.2, Based on the total impact vector The events are divided into the following three levels:

[0162] Low-level events: ;

[0163] Medium-level events: ;

[0164] High-level event: ;

[0165] S4.2.3, generating a feedback signal according to the event level , and feeding the feedback signal back to the energy management system;

[0166] ;

[0167] wherein, represents a low-level event; represents a medium-level event; represents a high-level event.

[0168] Embodiment 2:

[0169] The multi-source collaborative optimization scheduling method of the energy management system of the multi-energy complementary microgrid is used for the energy management system of the multi-energy complementary microgrid described above, and includes the following steps:

[0170] S8.1, calculating the device aging factor by applying an exponential decay model according to the cumulative use length of the device and the current load condition;

[0171] In this embodiment S8.1, the utilization efficiency, the basic allocation weight, and the load demand of each energy device in the energy management system of the multi-energy complementary microgrid are obtained, and the device aging factor is calculated by applying an exponential decay model according to the cumulative use length of the device and the current load condition. The specific method steps are as follows:

[0172] S8.1.1, obtaining the utilization efficiency of each energy in the energy management system of the multi-energy complementary microgrid , the basic allocation weight of each energy , the load demand , the output of each energy , and the feedback signal ;

[0173] S8.1.2, calculating the device aging factor by applying an exponential decay model according to the cumulative use length of the device and the current load condition :

[0174] ;

[0175] wherein, is the number variable of the energy device, , when takes , it represents the solar device aging factor, and so on; is the aging rate coefficient of the energy device; is the aging rate adjustment function; is the integral variable;

[0176] In this embodiment, the aging rate coefficient of the energy equipment is determined based on equipment life data, historical performance data, or aging rate information provided by the equipment manufacturer; the aging rate adjustment function is used to describe the aging rate of the equipment at a specific time point, reflecting the difference in the aging rate of the equipment under different load conditions. It can be adjusted according to the current load or external environment temperature and humidity, and set by empirical data or formula based on the running state of the equipment;

[0177] S8.2, construct an energy scheduling differential equation combined with the equipment aging factor, solve the energy scheduling differential equation, and obtain the energy correction weight;

[0178] In this embodiment S8.2, the energy scheduling differential equation is constructed combined with the equipment aging factor, the energy scheduling differential equation is solved, and the energy correction weight is obtained, the specific steps are as follows:

[0179] S8.2.1, construct an energy scheduling differential equation combined with the equipment aging factor :

[0180] ;

[0181] wherein, is the aging influence adjustment coefficient of the energy equipment; is the energy correction weight;

[0182] S8.2.2, solve the energy scheduling differential equation to obtain the energy correction weight .

[0183] S8.3, based on the energy correction weight and the feedback signal, introduce an adaptive weight mapping function to define the distribution strategy of the energy output, establish a distribution optimization probability control objective function through multiple probability control techniques, and use the Lagrange multiplier method to solve and obtain the optimal distribution weight;

[0184] In this embodiment S8.3, based on the energy correction weight and the feedback signal , an adaptive weight mapping function is introduced to define the distribution strategy of the energy output, a distribution optimization probability control objective function is established through multiple probability control techniques, and the Lagrange multiplier method is used to solve and obtain the optimal distribution weight, the specific method is as follows:

[0185] S8.3.1, construct a distribution weight smoothing nonlinear function according to the feedback signal :

[0186] ;

[0187] in, For event response sensitivity;

[0188] In this embodiment, event response sensitivity Used to adjust the influence strength of the feedback signal, the initial value is 1, and it is fine-tuned according to the debugging requirements of the system during actual operation;

[0189] S8.3.2, Energy-based weighting Construct an adaptive weight mapping function for each energy source:

[0190] ;

[0191] in, for Optimal weight allocation at any given time; This is the adjustment coefficient;

[0192] In this embodiment, the adjustment coefficient The value ranges from 0.1 to 1.0, and the optimal value is determined through simulation testing and optimization based on actual system operation data.

[0193] S8.3.3, based on Optimal weight allocation at any given time Energy utilization efficiency Weighting of each energy base Load demand Energy output and feedback signals Establish the allocation optimization probabilistic control objective function :

[0194] ;

[0195] in, To allocate and optimize the probabilistic control objective function; For adjustment terms;

[0196] S8.3.4 Constructing the Lagrange function The Lagrange multiplier method is used to solve the objective function of minimizing loss, thus obtaining the optimal weight allocation. :

[0197] ;

[0198] in, It is a Lagrange multiplier.

[0199] In the embodiment, the multi-source collaborative optimization scheduling method of the energy management system of the multi-energy complementary microgrid uses the method of combining the exponential decay model with the weighted differential equation to more finely integrate the device aging effect into the weight calculation and to affect the real-time adjustment of the energy distribution weight of each energy source, so that the system can dynamically optimize the energy distribution according to the aging degree of the device. After the aging effect is dynamically integrated into the distribution weight, the system can autonomously identify and reduce the dependence on the aging device without relying on manual intervention, thereby effectively avoiding the risk of failure caused by excessive load of the aging device.

[0200] The adaptive weight mapping and probability control optimization technology is a method of dynamically adjusting the resource allocation of the system, which optimizes the resource utilization efficiency and minimizes the system loss by real-time evaluation of the system state and the occurrence probability of events.

[0201] In the adaptive weight mapping, a variable weight is allocated to each resource or device according to the current state of the device, the environmental conditions and the system requirements, so as to allow the resource allocation weight to flexibly respond to the actual situation with the change of time and the aging degree of the device.

[0202] The probability control optimization evaluates the possibility of abnormal events by setting a probability model and adds this probability factor to the loss objective function. When the probability of abnormal events is high, the system will preferentially adjust the resource configuration to ensure the stability of the system in uncertain conditions.

[0203] By combining the adaptive weight mapping and the probability control optimization, more flexible energy distribution is realized.

[0204] S8.4, apply the optimal distribution weight to the energy management system of the multi-energy complementary microgrid to dynamically allocate the output of each energy source.

[0205] The above shows and describes the basic principles, main features and advantages of the present application. It should be understood by those skilled in the art that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application.

Claims

1. An energy management system for a multi-energy complementary microgrid, characterized in that, include: Energy monitoring unit (1), which is used to monitor the operation of equipment and energy consumption in the microgrid in real time; Energy management unit (2), which is based on differential geometric manifold theory and optimizes the allocation of each energy source by constructing a multidimensional manifold for energy flow; The load prediction unit (3) is used to monitor the equipment load, define and calculate the fault prediction index, predict equipment faults based on the fault prediction index, and dynamically reduce the load of faulty equipment. Anomaly detection unit (4) detects abnormal events based on the impact response matrix and provides hierarchical feedback of the abnormal events; The energy management unit (2) includes an energy multidimensional manifold construction module and an energy optimization allocation module; The energy multidimensional manifold construction module is based on differential geometric manifold theory to construct energy multidimensional manifolds. Generate manifold eigenvectors The specific steps are as follows: S2.1.1, Define solar power output as... Wind power output is The remaining energy storage capacity is Gas-fired power generation output is ; Solar energy utilization efficiency is defined as Wind energy utilization efficiency is Energy storage utilization efficiency is The utilization efficiency of gas-fired power generation is ; Define total load demand as , Assigning weights to solar energy Assigning weights to wind energy Assigning weights to energy storage Assign weights to gas-fired power generation, and satisfy the following conditions: ; S2.1.2, Construction Time State Point : ； in, for Solar power output at all times for Wind energy utilization efficiency at all times for Remaining energy storage capacity at all times for Gas-fired power generation output at all times; S2.1.3, State points at all times as time changes The collection constitutes an energy multidimensional manifold : ; in, The observation period; This is the start time of the time period; This is the end time of the time period; S2.1.4, Calculation Always putting in effort and manifold eigenvectors : ； ； in, The proportion of solar energy in total power output; The proportion of wind energy in total power output; The proportion of energy storage in total power output; This represents the proportion of gas-fired power generation in total power output.

2. The energy management system for a multi-energy complementary microgrid according to claim 1, characterized in that: The energy monitoring unit (1) includes a data acquisition module and an energy monitoring module; The data acquisition module uses sensors and communication technologies to collect real-time data on device operation and device status in the microgrid. Equipment operating data includes equipment load, equipment operating time, equipment current, equipment voltage, and equipment temperature; The energy monitoring module is used to receive and display equipment operating data and equipment operating status in real time.

3. The energy management system for a multi-energy complementary microgrid according to claim 2, characterized in that: The energy optimization allocation module is based on manifold feature vectors. An objective function for optimizing the total system loss is established, the energy allocation weights are solved using the Lagrange multiplier method, and energy is allocated based on these weights. The specific steps are as follows: S2.2.1, Based on total load demand Based on the efficiency of each energy source, establish an objective function to optimize the total system loss: ; in, for Total load demand at any given time; S2.2.2 Constructing the Lagrange function The allocation weights of each energy source are determined using the Lagrange multiplier method: ; S2.2.3 Obtain the optimal allocation weight for solar energy. Optimal allocation weight of wind energy Optimal allocation weights for energy storage Optimal allocation weights for gas-fired power generation Determine the output ratio of each energy source.

4. The energy management system for a multi-energy complementary microgrid according to claim 1, characterized in that: The load forecasting unit (3) includes a fault detection module and a load adjustment module; The fault detection module monitors the load status of the equipment and calculates the fault prediction index. And determine whether the equipment has a risk of failure. The specific steps are as follows: S3.1.1, Based on the current load of the equipment and maximum load Define and calculate the fault prediction index. : ； S3.1.2 Setting Fault Risk Thresholds Fault prediction index With fault risk threshold Comparison: like This triggers load adjustment and generates a load adjustment signal; like If so, no load adjustment is needed; The load adjustment module receives the load adjustment signal from the fault detection module and dynamically adjusts the load. The specific steps are as follows: S3.2.1, Based on the load adjustment coefficient Calculate the adjusted load : ; in, This is the load adjustment factor. ; S3.2.2, Adjust the load Applied to equipment.

5. The energy management system for a multi-energy complementary microgrid according to claim 1, characterized in that: The anomaly detection unit (4) includes an event recognition module and a hierarchical feedback module; The event recognition module identifies and analyzes abnormal events of the equipment by constructing an impact response matrix and generates event type information. The specific steps are as follows: S4.1.1, Define the microgrid, the first The device is The device operating status value at any given time is This includes equipment operating data, namely equipment load, equipment operating time, equipment current, equipment voltage, and equipment temperature; No. The normal state threshold for each device is: ;No. The device is deviation in time ; S4.1.2, based on the first The device is The device operating status value at any given time is and the The normal state threshold for each device is: Calculate the first The device is deviation in time : ； S4.1.3, Define the anomaly threshold If the first The device is deviation in time Greater than the abnormal threshold Then for the first Each device constructs an impact response matrix. : ； in, Impact response matrix The element in represents the first element. The device in abnormal condition affects the first The extent of the impact on each device; This is the proportionality coefficient; For the first The device in abnormal condition affects the first The influence weight of each device ; S4.1.4, Integrate the impact response matrix of all devices. Construct the total impact vector : 。 6. The energy management system for a multi-energy complementary microgrid according to claim 5, characterized in that: The graded feedback module is based on the total impact vector. The abnormal events will be categorized and reported in a tiered manner. The specific steps are as follows: S4.2.1 Define two event level thresholds and ; S4.2.2, Based on the total impact vector The events are divided into the following three levels: Low-level events: ; Medium-level events: ; High-level events: ; S4.2.3 Generate feedback signals based on event levels and will provide feedback signals Feedback is sent to the energy management system; ; in, Represents low-level events; Represents a medium-level event; This represents a high-level event.

7. A multi-source collaborative optimization scheduling method for the energy management system of a multi-energy complementary microgrid, used in the energy management system of the multi-energy complementary microgrid as described in any one of claims 1-6, characterized in that, Includes the following steps: S8.

1. Based on the cumulative usage time of the equipment and the current load conditions, calculate the equipment aging factor using the exponential decay model; S8.

2. Construct an energy scheduling differential equation by combining the equipment aging factor, solve the energy scheduling differential equation, and obtain the energy correction weight; S8.3 Based on energy correction weights and feedback signals, an adaptive weight mapping function is introduced to define the energy output allocation strategy. Through multiple probabilistic control techniques, an allocation optimization probabilistic control objective function is established, and the optimal allocation weights are obtained by solving the Lagrange multiplier method. S8.4 Apply the optimal allocation weight to the energy management system of the multi-energy complementary microgrid to dynamically allocate the output of each energy source.

8. The multi-source collaborative optimization scheduling method for the energy management system of a multi-energy complementary microgrid according to claim 7, characterized in that: In step S8.1, the utilization efficiency, basic allocation weight, and load demand of each energy device in the energy management system of the multi-energy complementary microgrid are obtained. Based on the cumulative usage time of the devices and current load conditions, an exponential decay model is applied to calculate the device aging factor. The specific steps are as follows: S8.1.1 Obtaining the energy utilization efficiency of each energy source in the energy management system of a multi-energy complementary microgrid. Weighting of each energy base Load demand Energy output and feedback signals ; S8.1.

2. Based on the cumulative usage time of the equipment and the current load conditions, calculate the equipment aging factor using the exponential decay model. : in, For the numbering variable of energy equipment, ,when Pick The time indicates the aging factor of the solar equipment, and so on; The aging rate coefficient of energy equipment; This is the aging rate adjustment function; For integration variables; In step S8.2, an energy scheduling differential equation is constructed by combining the equipment aging factor, and the energy scheduling differential equation is solved to obtain the energy correction weight. The specific steps are as follows: S8.2.1, combined with equipment aging factors Construct the energy scheduling differential equation: ； in, The adjustment coefficient for the aging impact of energy equipment; Adjust the weights for energy; S8.2.2 Solving the energy dispatch differential equation to obtain the energy correction weights .

9. The multi-source collaborative optimization scheduling method for the energy management system of a multi-energy complementary microgrid according to claim 8, characterized in that: In S8.3, based on energy correction weights and feedback signals An adaptive weight mapping function is introduced to define an energy output allocation strategy. Through multiple probabilistic control techniques, an allocation optimization probabilistic control objective function is established, and the optimal allocation weights are obtained using the Lagrange multiplier method. The specific method is as follows: S8.3.1, Based on the feedback signal Construct a weighted smoothing nonlinear function : ; in, For event response sensitivity; S8.3.2, Energy-based weighting Construct an adaptive weight mapping function for each energy source: ； in, for Optimal weight allocation at any given time; This is the adjustment coefficient; S8.3.3, based on Optimal weight allocation at any given time Energy utilization efficiency Weighting of each energy base Load demand Energy output and feedback signals Establish the allocation optimization probabilistic control objective function : ; in, To allocate and optimize the probabilistic control objective function; For adjustment terms; S8.3.4 Constructing the Lagrange function The Lagrange multiplier method is used to solve the objective function of minimizing loss, thus obtaining the optimal weight allocation. : ; in, It is a Lagrange multiplier.

Images