Diesel engine and battery economic operation control method fusing random impact load characteristics

Abstract

The invention relates to the technical field of energy management and control, in particular to a diesel engine and battery economic operation control method fusing random impact load characteristics, which comprises the following steps: predicting future random impact load based on a load prediction model; predicting a future long-term average load; determining a target SOC of the battery according to the future long-term average load; determining the output power of the diesel engine according to the target SOC, judging whether the determined output power of the diesel engine is in the optimal energy efficiency interval or not, if yes, adjusting the output power of the diesel engine according to the determined output power to ensure that the diesel engine always works in the optimal energy efficiency interval, and if not, judging whether the diesel engine needs to be shut down or not. Through the control method, under the dynamic load condition, the load change trend can be accurately predicted, meanwhile, it is ensured that the diesel engine always operates in the optimal energy efficiency interval, and therefore the economical efficiency of the whole system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management and control, and in particular to a diesel engine and battery economic operation control method integrating random impact load characteristics. Background Art

[0002] With the transformation of the global energy structure and the advancement of sustainable development needs, traditional energy systems are gradually transforming to clean and efficient hybrid systems. As the core device of traditional power systems, diesel engines still occupy an important position in many industries, especially in the fields of drilling rigs, construction machinery and off-grid power generation equipment with large power output requirements. However, load changes are inevitable in actual operation, which include not only smooth fluctuations, but also sudden load shocks (such as sudden increase or decrease in load). This may lead to reduced diesel engine operating efficiency, increased emissions, and inability to optimally schedule the energy flow between the diesel engine and the battery.

[0003] To solve this problem, batteries are gradually introduced into hybrid systems as energy storage devices to adjust load fluctuations and provide backup power. Batteries can absorb excess energy when the load drops suddenly and release energy when the load increases suddenly, thereby effectively improving the overall efficiency and economy of the system.

[0004] At present, although there are many studies and technologies trying to adjust load changes through batteries when load fluctuates, traditional control methods often ignore the prediction of future load trends, which leads to the failure to effectively optimize the operating state of the diesel engine, frequent battery charging and discharging times, and failure to achieve economical operation of the diesel engine and battery system.

[0005] In the prior art, a Chinese invention patent document with publication number CN117220318A and publication date December 12, 2023 is proposed to solve the above-mentioned technical problems. The technical solution disclosed in the patent document is as follows: a method and system for digital drive control of a power grid, comprising obtaining historical load information, historical weather information and photovoltaic power generation of a target power grid as original load characteristics, and determining the multi-layer perception characteristics corresponding to the original load characteristics through a feature extraction model of a preset load prediction model; predicting the load demand corresponding to the target power grid through a vector regression model of the load prediction model based on the multi-layer perception characteristics, wherein the load prediction model is constructed based on a deep learning model and a support vector regression model; obtaining the electric energy production of renewable energy in the target power grid in real time, and adaptively and dynamically adjusting the charging state and discharging state of the energy storage system in the target power grid according to the comparison result of the electric energy production and the load demand.

[0006] In actual use, the above technical solution can dynamically adjust the charging and discharging states of the energy storage system in the target power grid to a certain extent. However, the adjustment strategy is relatively simple and the following problems still exist: load volatility and random shocks will cause the diesel engine to frequently deviate from the optimal energy efficiency range, increase fuel consumption and emissions, and accelerate battery aging due to frequent charging and discharging. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes a diesel engine and battery economic operation control method that integrates random impact load characteristics. Under dynamic load conditions, it can accurately predict the load change trend, optimize the battery's target SOC setting and charge and discharge strategy to reduce the number of charge and discharge times, and ensure that the diesel engine always operates in the optimal energy efficiency range, thereby improving the economy of the entire system.

[0008] The present invention is achieved by adopting the following technical solutions:

[0009] The diesel engine and battery economic operation control method integrating random impact load characteristics comprises the following steps:

[0010] Step S 1 .Predict future random impact loads based on load forecasting models;

[0011] Step S 2 .Predict the future long-term average load;

[0012] Step S 3 .Determine the target SOC of the battery based on the future long-term average load;

[0013] Step S 4 .Determine the output power of the diesel engine according to the target SOC, and judge whether the determined output power of the diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power to ensure that the diesel engine always works within the optimal energy efficiency range. If not, judge whether the diesel engine needs to be shut down.

[0014] The method for determining the optimal energy efficiency range of a diesel engine is as follows: determining the optimal energy efficiency range of the diesel engine by utilizing a diesel engine power-speed curve and a fuel consumption rate-speed curve.

[0015] The method for determining the optimal energy efficiency range of a diesel engine specifically includes the following steps:

[0016] Determine the best operating speed range of the diesel engine according to the fuel consumption rate-speed curve;

[0017] Determine the second optimum operating speed range of the diesel engine according to the diesel engine power-speed curve;

[0018] According to the overlapping speed range of the best working speed range 1 and the best working speed range 2, the best energy efficiency range of the diesel engine [P min-opt , P max-opt ]]; where P min-opt is the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

[0019] Step S 1 The specific steps include:

[0020] Step S 11 . Obtain the historical load information of the diesel engine and the battery as the original load characteristics and extract the statistical characteristics of load fluctuations;

[0021] Step S 12 . Based on the load forecasting model, predict the load P at the i-th time point in the future future (t i );Wherein, the load forecasting model is constructed based on the periodic decomposition method and the LSTM algorithm model.

[0022] The step S 11 The specific steps include:

[0023] Step S 111 .Set up sensors to collect load data of diesel engine and battery in real time;

[0024] Step S 112 .Extract statistical characteristics of load fluctuations;

[0025] Step S 113 .Preprocess the statistical features, wherein the preprocessing includes filtering and outlier detection.

[0026] The step S 12 The load P at the i-th time point in the future future (t i )for:

[0027] P future (t i )=P trend (t i )+P seasonal (t i )+P residual (t i ),i={1,…,n},

[0028] In the formula, y i is a future time point, indicating the predicted time point y i =y+i*Δy; t is the current time point, used to input known historical data; P trend (yi ) is the long-term trend fitted by linear regression; P seasonal (y i ) is the periodic component estimated with period T; P residual (t i ) is the residual component, which is predicted by the LSTM algorithm model.

[0029] The step S 2 Specifically refers to:

[0030]

[0031] Where P long_term is the future long-term average load, P future (t i ) is the load at the i-th time point in the future; w i is the weight of the i-th time point, and λ is the time decay factor.

[0032] The step S 3 Specifically, it means gradually adjusting the target SOC of the battery according to the future long-term average load. The adjustment method is:

[0033] SOC target =SOC actual +β·(P long-term -P 0 (t))+α·(SOC base -SOC actual )+γ·σ p ,

[0034] In the formula, SOC target is the target SOC of the battery; SOC actual is the current real-time charging status of the battery; P long-term is the future long-term average load; P 0 (t) is the current load demand, (P long-term -P 0 (t)) represents the changing trend of future load demand; β is the trend adjustment coefficient; SOC base It refers to the static SOC when there is no significant change in load demand; α is the basic adjustment coefficient; σ p is the standard deviation of load fluctuation, and γ is the fluctuation adjustment coefficient.

[0035] Step S 4 The method for determining the output power of a diesel engine is:

[0036] P 1 (t+Δt)=max(P min-opt ,min(P max-opt , P 0(t)+k·(SOC target -SOC actual )),

[0037] Where P 0 (t) is the current load demand, P 1 (t+Δt) is the target load of the diesel engine at the next time point; k is the SOC deviation adjustment coefficient; SOC target is the target SOC of the battery, SOC actual is the current real-time charging status of the battery; P min-opt is the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

[0038] The method to judge whether the diesel engine needs to be shut down is: when P 1 (t+Δt) <P min-opt And SOC actual >

[0039] SOC min , it is determined that the diesel engine needs to stop running.

[0040] When the diesel engine is shut down, determine whether it needs to be restarted.

[0041] The method to judge whether the diesel engine needs to be restarted is: when P 1 (t+Δt)≥P min-opt or SOC actual ≤SOC min , it is judged that the diesel engine needs to be restarted.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] 1. The present invention dynamically adjusts the target SOC and the output power of the diesel engine through the future long-term average load. Specifically, the battery first provides power fluctuation smoothing for random impact loads to reduce diesel engine power fluctuations; then controls the charging and discharging trend of the battery according to the target SOC, optimizes the coordinated operation of the diesel engine and the battery, and adjusts the output power of the diesel engine to ensure that the diesel engine always works within the optimal energy efficiency range or stops working, avoids inefficient operation or overload, reduces the non-efficient operation of the diesel engine due to load fluctuations, and reduces fuel consumption and emissions.

[0044] Through the above scheme, the energy flow between the battery and the diesel engine can be reasonably scheduled to ensure that the system can operate stably under load fluctuations and shocks, thereby improving the overall economic benefits of the system.

[0045] 2. The present invention determines the optimal energy efficiency range of the diesel engine through the diesel engine power-speed curve and the fuel consumption rate-speed curve. The determination method is simple and intuitive, and the determination result is accurate.

[0046] 3. In the present invention, the adjustment of the target SOC is achieved by the combined effect of the three adjustment items: the adjustment item based on the load demand trend, the adjustment item based on the basic SOC, and the load fluctuation adjustment item. The adjustment item based on the load demand trend is the main part, which is adjusted based on future load demand; the adjustment item based on the basic SOC is an auxiliary item, which is used to avoid long-term overcharging or over-discharging of the battery and maintain the battery health; the load fluctuation adjustment item is an auxiliary item, which is used to increase the safety margin of the system and prevent the battery from being insufficient due to sudden load changes.

[0047] 4. The present invention also needs to determine whether it needs to be shut down and whether it needs to be restarted. Through this strategy, the safe operation of the system can be ensured, and the mutual coordination relationship between the battery and the diesel engine can be better. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, wherein:

[0049] Figure 1 A schematic diagram for determining the optimal energy efficiency range in the present invention;

[0050] Figure 2 It is a schematic diagram for comparing the predicted value and the actual value obtained by using this method in the present invention. DETAILED DESCRIPTION

[0051] Example 1

[0052] As a basic embodiment of the present invention, the present invention includes a diesel engine and battery economic operation control method integrating random impact load characteristics, comprising the following steps:

[0053] Step S 1 . Predict future random impact loads based on a load prediction model. The load prediction model can be predicted using an existing model, such as the model proposed in publication number CN117220318A, that is, the load prediction model is constructed based on a deep learning model and a support vector regression model.

[0054] Step S 2 .Predict the future long-term average load.

[0055] Step S 3 .Determine the target SOC of the battery based on the future long-term average load.

[0056] Step S 4.Determine the output power of the diesel engine according to the target SOC, and judge whether the determined output power of the diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power to ensure that the diesel engine always works within the optimal energy efficiency range; if not, judge whether the diesel engine needs to be shut down.

[0057] Example 2

[0058] As a preferred embodiment of the present invention, the present invention includes a diesel engine and battery economic operation control method integrating random impact load characteristics, comprising the following steps:

[0059] Step S 1 . Based on the load forecasting model, predict the future random impact load, that is, the load P at the i-th time point in the future future (y i ).

[0060] Step S 2 .Predict the future long-term average load P long_term :

[0061]

[0062] In the formula, w i is the weight of the i-th time point, and λ is the time decay factor.

[0063] Step S 3 .Determine the target SOC of the battery based on the future long-term average load:

[0064] SOC target =SOC actual +β·(P long-term -P 0 (t))+α·(SOC base -SOC actual )+γ·σ p ,

[0065] In the formula, SOC target is the target SOC of the battery, SOC actual is the current real-time charging status of the battery, P 0 (t) is the current load demand, (P long-term -P 0 (t)) represents the changing trend of future load demand; β is the trend adjustment coefficient; SOC base Refers to the static SOC when there is no significant load demand change; SOC base -SOC actual is the deviation between the current SOC and the basic SOC; α is the basic adjustment coefficient; σ p is the standard deviation of load fluctuation, and γ is the fluctuation adjustment coefficient.

[0066] Step S 4 .Determine the output power of the diesel engine according to the target SOC, and judge whether the determined output power of the diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power to ensure that the diesel engine always works within the optimal energy efficiency range; if not, judge whether the diesel engine needs to be shut down.

[0067] Among them, the optimal energy efficiency range of the diesel engine can be determined by using the diesel engine power-speed curve and fuel consumption rate-speed curve.

[0068] Example 3

[0069] As another preferred embodiment of the present invention, the present invention includes a diesel engine and battery economic operation control method integrating random impact load characteristics, comprising the following steps:

[0070] Step S 1 .Predict future random impact loads based on the load prediction model. Specifically, it includes the following steps:

[0071] Step S 11 . Obtain the historical load information of the diesel engine and the battery as the original load characteristics and extract the statistical characteristics of the load fluctuation.

[0072] Step S 12 . Based on the load forecasting model, predict the future random impact load, that is, the load P at the i-th time point in the future future (t i ). Wherein, the load forecasting model is constructed based on the periodic decomposition method and the LSTM algorithm model.

[0073] Step S 2 .Predict the future long-term average load.

[0074] Step S 3 .Determine the target SOC of the battery based on the future long-term average load.

[0075] Step S 4 .Determine the output power of the diesel engine according to the target SOC, and judge whether the determined output power of the diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power to ensure that the diesel engine always works within the optimal energy efficiency range. If not, judge whether the diesel engine needs to be shut down.

[0076] Among them, the method for determining the output power of the diesel engine is:

[0077] P 1 (t+Δt)=max(P min-opt ,min(P max-opt , P0 (t)+k·(SOC target -SOC actual )),

[0078] Where P 0 (t) is the current power of the diesel engine, P 1 (t+Δt) is the target power of the diesel engine at the next time point; k is the SOC deviation adjustment coefficient. target is the target SOC of the battery, SOC actual It is the current real-time charging status of the battery. target >SOC actual , the diesel engine increases power to charge the battery; when SOC target <SOC actual , the diesel engine reduces power, discharges the battery, and reduces diesel consumption; at the same time, it ensures that the diesel engine power always runs in the optimal energy efficiency range [P min-opt , P max-opt ].

[0079] Example 4

[0080] As the best implementation mode of the present invention, the present invention includes a diesel engine and battery economic operation control method integrating the characteristics of random impact loads. Specifically, the battery provides power fluctuation smoothing for the random impact load, so that the diesel engine always operates within the optimal energy efficiency range. When the load is relatively high, the battery discharges to provide power support; when the load is relatively low, the battery absorbs and stores the excess electricity generated by the diesel engine. Specifically, the following steps are included:

[0081] Step S 1 .Predict future random impact loads based on the load prediction model. Specifically, it includes the following steps:

[0082] Step S 11 . Obtain the historical load information of the diesel engine and the battery as the original load characteristics, and extract the statistical characteristics of the load fluctuation. More specifically, the following steps are included:

[0083] Step S 111 .Set up sensors to collect load data of diesel engine and battery system in real time. Specifically, the power output and speed status of diesel engine can be monitored through the joint action of current sensor, voltage sensor and speed sensor. The charging and discharging power and SOC of battery can be monitored in real time through BMS system.

[0084] Step S 112 . Extract statistical features of load fluctuations. The main features include historical data of power demand, equipment load fluctuations, and sudden increases or decreases in load.

[0085] Step S 113.Preprocess the statistical features, wherein the preprocessing includes filtering and outlier detection.

[0086] Step S 12 . Based on the load forecasting model, predict the load P at the i-th time point in the future future (t i ). The load forecasting model is constructed based on the periodic decomposition method and the LSTM algorithm model. The load P at the i-th time point in the future future (t i ) is determined by:

[0087] P future (t i )=P trend (t i )+P seasonal (t i )+P residual (t i ),i={1,…,n},

[0088] Where, t i is a future time point, indicating the predicted time point t i =t+i*Δt; t is the current time point, used to input known historical data; P trend (t i ) is the long-term trend fitted by linear regression; P seasonal (t i ) is the periodic component estimated with period T; P residual (t i ) is the residual component, which is predicted by the LSTM algorithm model.

[0089] Specifically, the least squares method is used to fit historical data to obtain the trend component P seasonal (t i )=S(t i modT).

[0090] In the formula, S(t i modT) is a periodic function that represents the fluctuation amplitude within a period T, where T is the length of the period (such as 24 hours), (t i modT) represents the future time point t i The position in the period T. By sliding the window to S(t i modT) to remove the noise that may be contained in the periodic value:

[0091]

[0092] Where W is the width of the smoothing window (taking the nearest W points for averaging), t i+k is the current cycle position t i Around the points (slide back and forth W / 2 points).

[0093] Step S 13 .Use LSTM algorithm to calculate the residual P residual (t i ) modeling and constructing the supervised learning sample input layer:

[0094] I t =[R(tn),R(t-n+1),…,R(t-1)],I t =R(t+1),

[0095] Among them, I t is the input feature, the residual sequence of the past n time steps.

[0096] Step S 14 .Build the LSTM hidden layer and capture the temporal dependency of the residuals through three gating mechanisms (forget gate, input gate, output gate):

[0097] Forget gate: f t =Sigmoid(w f *[h t-1 ,I t ]+b f ),

[0098] Input gate: i t =Sigmoid(w i *[h t-1 ,I t ]+b i ),

[0099] Output gate: o t =Sigmoid(w o *[h t-1 ,I t ]+b o ),

[0100] Status update: C t =f t *C t-1 +i t *tanh(w c *[h t-1 ,I t ]+b c ),

[0101] Hidden state: h t =o y *tanh(C t ).

[0102] Step S15 .Construct the LSTM output layer to predict the residual value of the next h time steps:

[0103]

[0104] Step S 16 The STL and LSTM combination method synthesizes the predicted trend, periodicity and residual to generate the final predicted value:

[0105]

[0106] Refer to the instruction manual Figure 2 , through this method, a more accurate prediction value can be obtained.

[0107] Step S 2 .Predict the future long-term average load:

[0108]

[0109] Where P long_term is the future long-term average load; P future (t i ) is the load at the i-th time point in the future; w i is the weight at the i-th time point; λ is the time decay factor, which controls the rate of weight decrease, and is usually in the range of 0<λ≤1. When λ is small, the weight decreases slowly and more evenly; when λ is large, the weight decreases quickly and is more concentrated at the most recent time point.

[0110] Step S 3 .Determine the target SOC of the battery based on the future long-term average load:

[0111] SOC target =SOC actual +β·(P long-term -P 0 (t))+α·(SOC base -SOC actual )+γ·σ p ,

[0112] In the formula, SOC target It is the target SOC of the battery, which is used to guide the charging and discharging trend of the battery. actual It is the current real-time charging status of the battery. As the basis for adjusting the target SOC, the target SOC needs to change gradually based on the current SOC, rather than suddenly.

[0113] The above formula is mainly divided into three parts: the first part is the adjustment item of load demand trend, and the target SOC is mainly adjusted based on future load demand. long-termIt is the future long-term average load, reflecting the future load trend of the system. 0 (t) is the current power demand, i.e. load demand. (P long-term -P 0 (t)) represents the changing trend of future load demand. β is the trend adjustment coefficient, which controls the response sensitivity of the target SOC to the change of load demand. long-term >P 0 (t): In the future, the load demand will increase and the target SOC will be increased to store more energy. long-term >P 0 (t): In the future, load demand decreases and the target SOC is lowered to reduce unnecessary reserves.

[0114] The second part is the basic SOC adjustment item, which is used to avoid long-term overcharging or over-discharging of the battery and maintain the battery health. base Refers to the static SOC when there is no significant change in load demand. It is the reference point for system operation and can be set to approximately 50% to 80% of the battery capacity to protect the battery and extend its life. base -SOC actual is the deviation between the current SOC and the base SOC. α is the base adjustment coefficient, which controls the speed at which the target SOC approaches the base value.

[0115] The third part is load fluctuation adjustment, γ·σ p Used to increase the safety margin of the system to prevent sudden load changes from causing insufficient battery energy (auxiliary). p is the standard deviation of load fluctuation, indicating the random volatility of load. γ is the fluctuation adjustment coefficient, which controls the buffer reserve capacity of the target SOC

[0116] If the load fluctuation is large (σ p >0): Target SOC is increased, and battery reserve is increased to cope with uncertainty. If the load fluctuation is small (σ p ≈0): The target SOC is mainly adjusted according to the load trend, without adding additional reserves.

[0117] Step S 4 .Determine the output power of the diesel engine according to the target SOC. Specifically, the battery first provides power fluctuation smoothing for random impact loads. Use the target SOC to guide the charging and discharging trend of the battery. If the actual SOC of the battery is less than the target SOC, increase the output power of the diesel generator; if the actual SOC of the battery is greater than the target SOC, reduce the output power of the diesel generator; if the actual SOC of the battery is near the target SOC, maintain the current output power of the diesel generator. The amplitude of adjusting the output power of the diesel generator is proportional to the deviation between the actual SOC of the battery and the target SOC. During the adjustment process, ensure that the diesel generator always operates within the optimal energy efficiency range or stops working.

[0118] The method for determining the optimal energy efficiency range of a diesel engine is as follows: determining the optimal energy efficiency range of a diesel engine by using a diesel engine power-speed curve and a fuel consumption rate-speed curve. Specifically, the following steps are included:

[0119] Determine the best working speed range 1 of the diesel engine based on the fuel consumption rate-speed curve. Determine the best working speed range 2 of the diesel engine based on the diesel engine power-speed curve. Determine the best energy efficiency range [P min-opt , P max-opt ]; where P min-opt is the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

[0120] For details, please refer to the attached manual. Figure 1 It can be seen that the fuel consumption rate presents a U-shaped curve, and the fuel consumption rate of the diesel engine is the lowest in the medium speed range of 1400rpm-1800rpm. At the same time, according to:

[0121] P=T·n,

[0122] In the formula, P is the output power of the diesel engine, T is the torque of the diesel engine, and n is the speed. The output power efficiency of the diesel engine is the highest in the medium speed range of 1200rpm-1800rpm. Combining the diesel engine power-speed curve and fuel consumption rate-speed curve, it can be known that the best energy efficiency range of the diesel engine is [140kw(P min-opt) , 170kw(P max-opt )].

[0123] More specifically, the method for determining the output power of a diesel engine is:

[0124] P 1 (t+Δt)=max(P min-opt ,min(P max-opt , P 0 (t)+k·(SOC target -SOC actual )),

[0125] Where P 0 (t) is the current power of the diesel engine, i.e. the current load demand; P 1 (t+Δt) is the target power of the diesel engine at the next time point, i.e., the target load; k is the SOC deviation adjustment coefficient, which determines the impact of the SOC deviation on the diesel engine power; SOC target is the target SOC of the battery; SOC actual is the current real-time charging status of the battery; P min-optis the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

[0126] Determine whether the output power of the determined diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power, and SOC target >SOC actual , the diesel engine increases power to charge the battery; when SOC target <SOC actual , the diesel engine reduces power, discharges the battery, and reduces diesel consumption. At the same time, it ensures that the diesel engine power always runs in the optimal energy efficiency range [P min-opt , P max-opt ] to avoid low-load inefficient operation or overload.

[0127] If not, determine whether the diesel engine needs to be shut down. The method for determining whether the diesel engine needs to be shut down is: when P 1 (t+Δt) <P min-opt And SOC actual >SOC min , it is determined that the diesel engine needs to stop running.

[0128] SOC min It is the minimum safe SOC of the battery to prevent over-discharge (such as 20%-30%).

[0129] The shutdown logic needs to meet two conditions at the same time:

[0130] First, the load is low: if the load is lower than the minimum efficient load P of the diesel engine min-opt , the diesel engine's operating efficiency is low, and battery power should be used first.

[0131] Second, the battery is sufficient: If the SOC is higher than the safety lower limit SOC min , the battery can continue to provide power without the need for the diesel engine to intervene.

[0132] When the diesel engine is shut down, determine whether it needs to be restarted.

[0133] The method to judge whether the diesel engine needs to be restarted is: when P 1 (t+Δt)≥P min-opt or SOC actual ≤SOC min , it is judged that the diesel engine needs to be restarted.

[0134] The restart logic only needs to meet one of the following two conditions:

[0135] First, when the load is higher than the diesel engine's lowest efficient load, the diesel engine starts: If the current load is high and the battery cannot supply power independently, the diesel engine restarts to provide power.

[0136] Second, when the battery SOC is lower than the minimum limit, the diesel engine starts: If the battery SOC drops to SOC min , the diesel engine starts and charges the battery to prevent the battery from being damaged by over-discharge.

[0137] In summary, after reading the present invention document, ordinary technicians in this field can make various other corresponding transformation schemes based on the technical scheme and technical concept of the present invention without creative mental labor, which all fall within the scope of protection of the present invention.

Claims

1. A diesel engine and battery economic operation control method integrating random impact load characteristics, characterized by: The following steps are involved: Step S1. Predicting future random impact loads based on a load prediction model; Step S2. Predicting the future long-term average load; Step S3. Determine the target SOC of the battery according to the future long-term average load; Step S4. Determine the output power of the diesel engine according to the target SOC, and judge whether the determined output power of the diesel engine is within the optimal energy efficiency range. If so, adjust the output power of the diesel engine according to the determined output power to ensure that the diesel engine always operates within the optimal energy efficiency range. If not, judge whether the diesel engine needs to be shut down.

2. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 1 is characterized in that: The method for determining the optimal energy efficiency range of a diesel engine is as follows: determining the optimal energy efficiency range of the diesel engine by utilizing a diesel engine power-speed curve and a fuel consumption rate-speed curve.

3. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 2 is characterized in that: The method for determining the optimal energy efficiency range of a diesel engine specifically includes the following steps: Determine the best operating speed range of the diesel engine according to the fuel consumption rate-speed curve; Determine the second optimum operating speed range of the diesel engine according to the diesel engine power-speed curve; According to the overlapping speed range of the best working speed range 1 and the best working speed range 2, the best energy efficiency range of the diesel engine [P min-opt , P max-opt ]; where P min-opt is the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

4. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 1 is characterized in that: Step S1 specifically includes the following steps: Step S 11 . Obtain the historical load information of the diesel engine and the battery as the original load characteristics and extract the statistical characteristics of load fluctuations; Step S 12 . Based on the load forecasting model, predict the load P at the i-th time point in the future future (t i );Wherein, the load forecasting model is constructed based on the periodic decomposition method and the LSTM algorithm model.

5. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 4 is characterized in that: The step S 11 The specific steps include: Step S 111 .Set up sensors to collect load data of diesel engine and battery in real time; Step S 112 .Extract statistical characteristics of load fluctuations; Step S 113 .Preprocess the statistical features, wherein the preprocessing includes filtering and outlier detection.

6. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 4 is characterized in that: The step S 12 The load P at the i-th time point in the future future (t i )for: P future (t i )=P trend (t i )+P seasonal (t i )+P residual (t i ),i={1,…,n}, Where, t i is a future time point, indicating the predicted time point t i =t+i*Δt; t is the current time point, used to input known historical data; P trend (t i ) is the long-term trend fitted by linear regression; P seasonal (t i ) is the periodic component estimated with period T; P residual (t i ) is the residual component, which is predicted by the LSTM algorithm model.

7. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 1 is characterized in that: The step S2 specifically refers to: Where P long_term is the future long-term average load, P future (t i ) is the load at the i-th time point in the future; w i is the weight of the i-th time point, and λ is the time decay factor.

8. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 1 is characterized in that: The step S3 specifically refers to: gradually adjusting the target SOC of the battery according to the future long-term average load, and the adjustment method is: SOCIETY target =SOC actual +β·(P long-term -P0(t))+α·(SOC base -SOC actual )+γ·σ p , In the formula, SOC target is the target SOC of the battery; SOC actual is the current real-time charging status of the battery; P long-term is the future long-term average load; P0(t) is the current load demand, (P long-term -P0(t)) represents the changing trend of future load demand; β is the trend adjustment coefficient; SOC base It refers to the static SOC when there is no significant load demand change; α is the basic adjustment coefficient; σ p is the standard deviation of load fluctuation, and γ is the fluctuation adjustment coefficient.

9. The diesel engine and battery economic operation control method integrating random impact load characteristics according to any one of claims 1 to 8, characterized in that: The method for determining the output power of the diesel engine in step S4 is: P1(t+Δt)=max(P min-opt ,min(P max-opt , P0(t)+k·(SOC target -SOC actual )), where P0(t) is the current load demand, P1(t+Δt) is the target load of the diesel engine at the next time point; k is the SOC deviation adjustment coefficient; SOC target is the target SOC of the battery, SOC actual is the current real-time charging status of the battery; P min-opt is the lowest efficient load of the diesel engine, P max-opt It is the highest efficient load of the diesel engine.

10. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 9 is characterized in that: The method to judge whether the diesel engine needs to be shut down is: when P1(t+Δt) <P min-opt And SOC actual >SOC min , it is determined that the diesel engine needs to stop running.

11. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 10 is characterized in that: When the diesel engine is shut down, determine whether it needs to be restarted.

12. The diesel engine and battery economic operation control method integrating random impact load characteristics according to claim 11 is characterized in that: The method to determine whether the diesel engine needs to be restarted is: when P1(t+Δt)≥P min-opt or SOC actual ≤SOC min , it is judged that the diesel engine needs to be restarted.

Images