A sewage treatment water quality prediction method and system

Abstract

The present invention discloses a method and system for predicting the water quality of sewage treatment, belonging to the technical field of sewage treatment. The present invention reflects the pollution situation of the influent by collecting the sewage influent flow rate and water quality indicators in real time, calculates the sewage residence time according to the sewage influent flow rate at each moment, compares the sewage residence time with the reference time to construct a residence time deviation sequence, constructs an aeration volume deviation sequence according to the gap between the aeration volume provided historically and the required aeration volume, reflects the evaluation of the adaptability to water quality treatment according to the deviations of the residence time and the aeration volume, and at the same time, the change of the water quality indicator reflects the difficulty of water quality treatment during this period. The water quality treatment fitness score is obtained by combining the sewage water quality treatment process and the water quality treatment difficulty, which is used to evaluate the influent within this period of time, predict the water quality treatment fitness score in advance, so as to warn the staff to adjust the parameters of sewage treatment in time.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and particularly to a method and system for predicting the water quality of sewage treatment. Background Art

[0002] Sewage treatment is an important part of modern urban environmental management. With the acceleration of urbanization and the advancement of industrial development, the sewage discharge has increased significantly, and the water quality problem has become increasingly serious. To protect water resources and improve water quality, a series of sewage treatment measures have been widely adopted globally. These measures not only include physical and chemical treatment means, but also emphasize the importance of biological treatment in removing organic matter and nutrients in sewage.

[0003] Secondary biological treatment is one of the key links in sewage treatment. Through the metabolic activities of microorganisms, the chemical oxygen demand (COD) and biological oxygen demand (BOD) in sewage can be effectively reduced, thus meeting the discharge standards. Although the water quality indicators in the sewage treatment process can be directly detected, adjusting the sewage treatment parameters based on the real-time water quality indicators will have certain hysteresis and uncertainty, and at the same time, it will also extend the sewage treatment time and cannot predict the adaptability of the current water quality treatment process. Summary of the Invention

[0004] Aiming at the above deficiencies in the prior art, the method and system for predicting the water quality of sewage treatment provided by the present invention solve the problem that the prior art cannot predict the adaptability of the current water quality treatment process.

[0005] In order to achieve the above invention object, the technical solution adopted by the present invention is: A method for predicting the water quality of sewage treatment, comprising the following steps:

[0006] S1. Collect the sewage influent flow rate and influent water quality indicators at each moment, wherein the influent water quality indicators include: influent chemical oxygen demand, influent biological oxygen demand, and influent ammonia nitrogen concentration;

[0007] S2. Obtain multiple sewage residence times in the biochemical pool according to the sewage influent flow rate at each moment;

[0008] S3. Construct a residence time deviation sequence according to the difference between the multiple sewage residence times and the reference time;

[0009] S4. Construct an aeration volume deviation sequence according to the difference between the aeration volume provided at the historical moment and the required aeration volume;

[0010] S5. Construct a water quality indicator change sequence according to the influent water quality indicators at each moment;

[0011] S6. Predict the water quality treatment fitness score based on the residence time deviation sequence, the aeration volume deviation sequence, and the water quality indicator change sequence, based on the water quality prediction model.

[0012] Further, the reference time in S3 includes: the minimum residence time T min , the maximum residence time T max and the standard residence time T s . S3 includes the following sub-steps:

[0013] S31. When (T min + T t ) / 2 ≤ T t ≤ (T max + T t ), assign a time deviation coefficient of 0 to the corresponding sewage residence time T t ;

[0014] S32. When T min ≤ T t < (T min + T s ), calculate the time deviation coefficient of the corresponding sewage residence time T t : , where θ is the time deviation coefficient;

[0015] S33. When (T max + T s ) / 2 < T t ≤ T max , calculate the time deviation coefficient of the corresponding sewage residence time T t : ;

[0016] S34. When T t < T min , calculate the time deviation coefficient of the corresponding sewage residence time T t : ;

[0017] S35. When T t > T max , calculate the time deviation coefficient of the corresponding sewage residence time T t : ;

[0018] S36. Arrange each time deviation coefficient in the order of occurrence of the sewage residence time T t to obtain a residence time deviation sequence.

[0019] Further, S4 includes the following sub-steps:

[0020] S41. Calculate the required aeration volume for the influent water quality index at each moment to obtain a required aeration volume sequence;

[0021] S42. Obtain a comparison aeration volume sequence based on the aeration volumes provided at each historical moment;

[0022] S43. Subtract the comparison aeration volume sequence from the required aeration volume sequence to obtain an aeration volume difference sequence;

[0023] S44. Calculate an aeration volume deviation coefficient based on each aeration volume difference in the aeration volume difference sequence, and construct an aeration volume deviation sequence.

[0024] Further, the formula for calculating the aeration volume deviation coefficient in S44 is: , where μ is the aeration volume deviation coefficient, △Q a is the aeration volume difference, and | | represents the absolute value.

[0025] Further, S5 includes the following sub-steps:

[0026] S51. Subtract the influent water quality indicators at adjacent moments to obtain a water quality indicator difference;

[0027] S52. Calculate a water quality indicator change coefficient based on the water quality indicator difference, and construct a water quality indicator change sequence.

[0028] Further, the water quality indicator difference in S51 is: , where △C t is the water quality indicator difference at the t-th moment, C COD,in,t+1 is the influent chemical oxygen demand at the (t + 1)-th moment, C BOD,in,t+1 is the influent biochemical oxygen demand at the (t + 1)-th moment, C NH,in,t+1 is the influent ammonia nitrogen concentration at the (t + 1)-th moment, C COD,in,t is the influent chemical oxygen demand at the t-th moment, C BOD,in,t is the influent biochemical oxygen demand at the t-th moment, C NH,in,t is the influent ammonia nitrogen concentration at the t-th moment, and t is the number of the moment.

[0029] Further, the formula for calculating the water quality indicator change coefficient in S52 is: , where ε is the water quality indicator change coefficient, and | | represents the absolute value operation.

[0030] Further, the water quality prediction model in S6 includes: a time deviation processing channel, an aeration volume deviation processing channel, a water quality indicator change processing channel, and an output layer;

[0031] The input end of the time deviation processing channel is used to input the residence time deviation sequence to obtain the time influence value; the input end of the aeration volume deviation processing channel is used to input the aeration volume deviation sequence to obtain the aeration volume influence value; the input end of the water quality index change processing channel is used to input the water quality index change sequence to obtain the water quality index influence value; the output layer is used to weight the time influence value, the aeration volume influence value and the water quality index influence value to obtain the water quality treatment fitness score.

[0032] Further, the time deviation processing channel, the aeration volume deviation processing channel and the water quality index change processing channel all include, connected in sequence: a maximum pooling layer, a significant value enhancement layer and a fully connected layer, wherein the maximum pooling layer is a one-dimensional pooling layer, the pooling window size is 4, and the stride is 4;

[0033] The expression of the significant value enhancement layer is: , where y i is the i-th enhanced significant value output by the significant value enhancement layer, x i is the i-th significant value output by the maximum pooling layer, and e is the natural constant.

[0034] A sewage treatment water quality prediction system includes: a collection unit, a sewage residence time acquisition unit, a first sequence construction unit, a second sequence construction unit, a third sequence construction unit and a prediction unit;

[0035] The collection unit is used to collect the sewage influent flow rate and the influent water quality index at each moment;

[0036] The sewage residence time acquisition unit is used to obtain multiple sewage residence times in the biochemical tank according to the sewage influent flow rate at each moment;

[0037] The first sequence construction unit is used to construct a residence time deviation sequence according to the gap between the multiple sewage residence times and the reference time;

[0038] The second sequence construction unit is used to construct an aeration volume deviation sequence according to the gap between the aeration volume provided at the historical moment and the required aeration volume;

[0039] The third sequence construction unit is used to construct a water quality index change sequence according to the influent water quality index at each moment;

[0040] The prediction unit is used to predict the water quality treatment fitness score based on the residence time deviation sequence, the aeration volume deviation sequence and the water quality index change sequence according to the water quality prediction model.

[0041] The beneficial effects of the present invention are as follows: By collecting the sewage influent flow rate and water quality indicators (including chemical oxygen demand, biological oxygen demand, and ammonia nitrogen concentration) in real time, the present invention effectively reflects the pollution situation of the influent. According to the sewage influent flow rate at each moment, the residence time of the sewage in the biochemical pool is calculated and compared with the reference time to construct a residence time deviation sequence, which can reveal the adaptability of the sewage treatment in terms of time.

[0042] Meanwhile, the present invention also considers the gap between the actual aeration volume provided and the required aeration volume during the sewage treatment process, and constructs an aeration volume deviation sequence. The establishment of this sequence helps to evaluate the impact of aeration conditions on the sewage treatment effect.

[0043] In addition, the change of water quality indicators reflects the difficulty of water quality treatment. The rapidly changing water quality indicators may lead to an increase in treatment difficulty and affect the treatment effect. By combining the residence time deviation, aeration volume deviation, and water quality indicator changes, the adaptability of the sewage treatment is comprehensively evaluated, and finally a water quality treatment fitness score is obtained. The water quality treatment fitness score is used to evaluate the influent water during this period of time, predicting in advance the fitness situation of the water quality treatment process, so as to warn the staff to adjust the sewage treatment parameters in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flowchart of a sewage treatment water quality prediction method;

[0045] Figure 2 is a schematic structural diagram of a water quality prediction model. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0047] Example 1, as Figure 1 shown, a sewage treatment water quality prediction method includes the following steps:

[0048] S1. Collect the sewage influent flow rate and influent water quality indicators at each moment. Among them, the influent water quality indicators include: influent chemical oxygen demand, influent biological oxygen demand, and influent ammonia nitrogen concentration;

[0049] S2. Obtain multiple sewage residence times in the biochemical pool according to the sewage influent flow rate at each moment;

[0050] S3. Construct a residence time deviation sequence based on the differences between multiple sewage residence times and the reference time;

[0051] S4. Construct an aeration volume deviation sequence based on the difference between the aeration volume provided at historical moments and the required aeration volume;

[0052] S5. Construct a water quality index change sequence based on the influent water quality indexes at each moment;

[0053] S6. Predict the water quality treatment fitness score based on the residence time deviation sequence, the aeration volume deviation sequence, and the water quality index change sequence using a water quality prediction model.

[0054] In this embodiment, the calculation formula for the sewage residence time in the biochemical pool in S2 is: , where T t is the sewage residence time corresponding to the sewage influent flow rate at the t-th moment, Q t is the sewage influent flow rate at the t-th moment, and V is the volume of the biochemical pool.

[0055] In this embodiment, the reference time in S3 includes: the minimum residence time T min , the maximum residence time T max , and the standard residence time T s , and S3 includes the following sub-steps:

[0056] S31. When (T min +T t ) / 2 ≤ T t ≤ (T max +T t ) / 2, assign a time deviation coefficient of 0 to the corresponding sewage residence time T t ;

[0057] S32. When T min ≤ T t <(T min +T s ) / 2, calculate the time deviation coefficient of the corresponding sewage residence time T t : , where θ is the time deviation coefficient;

[0058] S33. When (T max +T s ) / 2 < T t ≤ T max , calculate the time deviation coefficient of the corresponding sewage residence time T t : ;

[0059] S34. When T t < T minWhen calculating the corresponding sewage retention time T t Time deviation coefficient: ;

[0060] S35. When T t > T max Calculate the time deviation coefficient of the corresponding sewage retention time T t : ;

[0061] S36. Arrange each time deviation coefficient in the order of occurrence of the sewage retention time T t To obtain the retention time deviation sequence.

[0062] In the secondary treatment process of the sewage treatment plant: the minimum retention time T min = 4 hours, the maximum retention time T max = 8 hours, and the standard retention time T s = 6 hours.

[0063] 1. When the actual retention time T t is within the ideal range (5 - 7 hours): when T t = 6 hours, at this time θ = 0, indicating that it is within the optimal retention time range;

[0064] 2. When the retention time is slightly short (4 - 5 hours): when T t = 4.5 hours, θ = 0.4, indicating a slight time shortage;

[0065] 3. When the retention time is slightly long (7 - 8 hours): T t = 7.5 hours, θ = 0.4, indicating a slight time over - length;

[0066] 4. When the retention time is severely insufficient (<4 hours): T t = 3 hours, θ = 0.67, indicating a relatively severe time shortage;

[0067] 5. When the retention time is too long (>8 hours): T t = 9 hours, θ = 0.83, indicating a relatively severe time over - length.

[0068] During the sewage treatment process, too short a retention time may cause the organic matter and nutrients in the sewage not to be fully degraded, while too long a retention time may lead to a reduction in sewage treatment efficiency and the risk of secondary pollution. In the present invention, for the retention time within the ideal range, an optimal evaluation (θ = 0) is given, for slightly deviated cases, a smaller deviation coefficient is given, and for severely deviated cases, a larger deviation coefficient is given.

[0069] In this embodiment, S4 includes the following sub-steps:

[0070] S41. Calculate the required aeration volume for the influent water quality index at each moment to obtain a sequence of required aeration volumes;

[0071] S42. Obtain a sequence of comparison aeration volumes based on the aeration volumes provided at each historical moment;

[0072] S43. Subtract the sequence of comparison aeration volumes from the sequence of required aeration volumes to obtain a sequence of aeration volume differences;

[0073] S44. Calculate the aeration volume deviation coefficient according to each aeration volume difference in the sequence of aeration volume differences, and construct a sequence of aeration volume deviations.

[0074] In this embodiment, the aeration volume data provided in S42 is sourced from: the aeration volume provided to the sewage during the sewage treatment process in a recent historical period, i.e., the historical aeration volume adjacent to the current one.

[0075] In this embodiment, the formula for calculating the required aeration volume in S41 is: , where Q a is the required aeration volume, C COD,in,t is the influent chemical oxygen demand at the t-th moment, C BOD,in,t is the influent biochemical oxygen demand at the t-th moment, C NH,in,t is the influent ammonia nitrogen concentration at the t-th moment, t is the number of the moment, and K1 is the proportionality constant of the influent chemical oxygen demand C COD,in,t is the proportionality constant of the influent biochemical oxygen demand C BOD,in,t is the proportionality constant of the influent ammonia nitrogen concentration C NH,in,t is the proportionality constant.

[0076] In this embodiment, other linear regression models can also be used to calculate the required aeration volume, and K1, K2, and K3 can be determined through experiments or experience.

[0077] In this embodiment, the formula for calculating the aeration volume deviation coefficient in S44 is: , where μ is the aeration volume deviation coefficient, △Q a is the aeration volume difference, and | | is the absolute value.

[0078] In the present invention, when the aeration volume difference is greater than or equal to 0, the provided aeration volume is sufficient. When the aeration volume difference is less than 0, the smaller the aeration volume difference, the greater the aeration volume deviation coefficient, and the greater the actually required aeration volume.

[0079] In this embodiment, S5 includes the following sub-steps:

[0080] S51. Subtract the influent water quality indexes at adjacent moments to obtain a water quality index difference;

[0081] S52. Calculate the water quality index change coefficient according to the difference in water quality indexes, and construct a water quality index change sequence.

[0082] In this embodiment, the difference in water quality indexes in S51 is: , where ΔC t is the difference in water quality indexes at the t-th moment, C COD,in,t+1 is the influent chemical oxygen demand at the (t + 1)-th moment, C BOD,in,t+1 is the influent biochemical oxygen demand at the (t + 1)-th moment, C NH,in,t+1 is the influent ammonia nitrogen concentration at the (t + 1)-th moment, C COD,in,t is the influent chemical oxygen demand at the t-th moment, C BOD,in,t is the influent biochemical oxygen demand at the t-th moment, C NH,in,t is the influent ammonia nitrogen concentration at the t-th moment, and t is the number of the moment.

[0083] In this embodiment, the formula for calculating the water quality index change coefficient in S52 is: , where ε is the water quality index change coefficient, and | | is the absolute value operation.

[0084] The present invention subtracts the influent water quality indexes at adjacent moments to obtain the difference in water quality indexes, which reflects the change amplitude of the water quality indexes, thereby calculating the water quality index change coefficient, and reflecting whether there is a sudden change in the influent water quality in a short time through the water quality index change coefficient.

[0085] When the water quality indexes suddenly increase, it will cause a drastic change in the growth environment of microorganisms, which may lead to the overgrowth of some microbial populations, while other populations may decrease, thus affecting the overall efficiency of sewage treatment. The treatment facilities cannot adapt to the sudden water quality change in time, resulting in insufficient treatment capacity, unqualified effluent water quality, and increasing the pollution risk to the environment. When the water quality indexes suddenly decrease, it means a reduction in pollutants in the influent water, which may lead to a decrease in the activity of microorganisms, affecting their treatment capacity for the remaining pollutants, and thus reducing the overall treatment efficiency. For a sewage treatment system, both a sudden large increase and a sudden large decrease in water quality indexes will increase the difficulty of sewage treatment.

[0086] In this embodiment, the water quality prediction model can adopt a BP neural network. Input the residence time deviation sequence, aeration volume deviation sequence, and water quality index change sequence into the input end of the BP neural network to obtain a water quality treatment fitness score. More preferably, as Figure 2 shown, the water quality prediction model in S6 includes: a time deviation processing channel, an aeration volume deviation processing channel, a water quality index change processing channel, and an output layer;

[0087] The input end of the time deviation processing channel is used to input the residence time deviation sequence to obtain the time influence value; the input end of the aeration volume deviation processing channel is used to input the aeration volume deviation sequence to obtain the aeration volume influence value; the input end of the water quality index change processing channel is used to input the water quality index change sequence to obtain the water quality index influence value; the output layer is used to weight the time influence value, the aeration volume influence value and the water quality index influence value to obtain the water treatment fitness score.

[0088] The time deviation processing channel calculates the influence of time on the treatment effect by receiving the residence time deviation sequence; the aeration volume deviation processing channel evaluates the effect of oxygen supply on microbial activities by analyzing the deviation of the aeration volume; while the water quality index change processing channel focuses on the dynamic impact of water quality changes. Finally, these influence values are weighted and synthesized in the output layer to form the water treatment fitness score.

[0089] In the output layer, corresponding weights are assigned to the time influence value, the aeration volume influence value and the water quality index influence value to obtain the water treatment fitness score.

[0090] The time deviation processing channel, the aeration volume deviation processing channel and the water quality index change processing channel all include, connected in sequence: a max pooling layer, a significant value enhancement layer and a fully connected layer, where the max pooling layer is a one-dimensional pooling layer, the pooling window size is 4, and the stride is 4;

[0091] The expression of the significant value enhancement layer is: , where y i is the i-th enhanced significant value output by the significant value enhancement layer, x i is the i-th significant value output by the max pooling layer, and e is the natural constant.

[0092] Since in the residence time deviation sequence of the present invention, the larger the time deviation coefficient, the more serious the time deviation; in the aeration volume deviation sequence, the larger the aeration volume deviation coefficient, the more serious the aeration volume deviation; in the water quality index change sequence, the larger the water quality index change coefficient, the more likely there is a mutation in the water quality index. Therefore, in the present invention, a max pooling layer with a size of 4 and a stride of 4 extracts a maximum value (i.e., the significant value) from every 4 moments in the sequence, and the most significant change situation is reflected by the maximum value, filtering out noise and unimportant fluctuations, and then using the significant value enhancement layer to enhance the significant value, so that the influence value can better reflect the deviation or change situation.

[0093] In this embodiment, the fully connected layer is used to process the enhanced significant values output by the significant value enhancement layer, and the activation functions adopted by the fully connected layer include: the Sigmoid function and the tanh activation function.

[0094] In this embodiment, the water quality prediction model can be trained by the gradient descent method.

[0095] Embodiment 2. A sewage treatment water quality prediction system includes: a collection unit, a sewage residence time acquisition unit, a first sequence construction unit, a second sequence construction unit, a third sequence construction unit, and a prediction unit;

[0096] The collection unit is used to collect the sewage inflow rate and influent water quality indicators at each moment;

[0097] The sewage residence time acquisition unit is used to obtain multiple sewage residence times in the biochemical tank according to the sewage inflow rate at each moment;

[0098] The first sequence construction unit is used to construct a residence time deviation sequence according to the gap between multiple sewage residence times and the reference time;

[0099] The second sequence construction unit is used to construct an aeration volume deviation sequence according to the gap between the aeration volume provided at the historical moment and the required aeration volume;

[0100] The third sequence construction unit is used to construct a water quality index change sequence according to the influent water quality indicators at each moment;

[0101] The prediction unit is used to predict the water quality treatment fitness score based on the residence time deviation sequence, the aeration volume deviation sequence, and the water quality index change sequence, based on the water quality prediction model.

[0102] The specific implementation process of Embodiment 2 is the same as that of Embodiment 1.

[0103] In this Embodiment 1 and 2, the higher the water quality treatment fitness score, the worse the existing sewage treatment situation. According to the water quality treatment fitness score, the following specific guiding suggestions can be provided for the staff: score intervals and corresponding measures, high fitness (1 - 2 points): The system is in good operating condition, maintain the current operating parameters; medium - high fitness (2 - 3 points): Check whether the residence time is within the optimal range, fine - tune the aeration volume to ensure that the dissolved oxygen is maintained at the ideal level; medium fitness (3 - 4 points): Adjust the influent flow rate to make the residence time closer to the reference value, appropriately increase or decrease the aeration volume according to actual needs; low fitness (4 - 5 points): Immediately check whether there is a mutation in the influent water quality indicators, adjust the operating parameters of the biochemical tank, including the reflux ratio and sludge concentration, increase the aeration volume, improve the treatment capacity, consider starting an emergency treatment plan, increase the monitoring frequency, and detect key indicators every 2 hours; extremely low fitness (5 points): Start the emergency plan and temporarily reduce the influent load.

[0104] The present invention effectively reflects the pollution situation of the influent by collecting the sewage influent flow rate and water quality indicators (including chemical oxygen demand, biological oxygen demand, and ammonia nitrogen concentration) in real time. According to the sewage influent flow rate at each moment, the residence time of the sewage in the biochemical tank is calculated and compared with the reference time to construct a residence time deviation sequence, which can reveal the adaptability of the sewage treatment time.

[0105] Meanwhile, the present invention also considers the gap between the actually provided aeration volume and the required aeration volume during the sewage treatment process, and constructs an aeration volume deviation sequence. The establishment of this sequence helps to evaluate the influence of aeration conditions on the sewage treatment effect.

[0106] In addition, the change of water quality indicators reflects the difficulty of water quality treatment. The rapidly changing water quality indicators may lead to an increase in treatment difficulty and affect the treatment effect. By combining the residence time deviation, aeration volume deviation, and water quality indicator change, the adaptability of the sewage treatment is comprehensively evaluated, and finally the water quality treatment fitness score is obtained. The water quality treatment fitness score is used to evaluate the influent during this period of time, predict in advance the fitness of the water quality treatment process, and thus warn the staff to adjust the sewage treatment parameters in a timely manner.

[0107] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for predicting the water quality of sewage treatment, characterized in that, It includes the following steps: S1. Collect the sewage influent flow rate and influent water quality indicators at each moment. Among them, the influent water quality indicators include: influent chemical oxygen demand, influent biological oxygen demand, and influent ammonia nitrogen concentration; S2. Obtain multiple sewage residence times in the biochemical pool according to the sewage influent flow rate at each moment; S3. Construct a residence time deviation sequence according to the gap between the multiple sewage residence times and the reference time; S4. Calculate the deviation coefficient of the gas injection volume according to the difference between the gas injection volume provided at the historical moment and the required gas injection volume, and construct a deviation sequence of the gas injection volume, , where μ is the deviation coefficient of the gas injection volume, and △Q a is the difference in the gas injection volume, and | | represents the absolute value; S5. Subtract the water quality index at adjacent times to obtain the water quality index difference: , where △C t is the water quality index difference at the t-th moment, C COD,in,t+1 is the influent chemical oxygen demand at the (t + 1)-th moment, C BOD,in,t+1 is the influent biochemical oxygen demand at the (t + 1)-th moment, C NH,in,t+1 is the influent ammonia nitrogen concentration at the (t + 1)-th moment, C COD,in,t is the influent chemical oxygen demand at the t-th moment, C BOD,in,t is the influent biochemical oxygen demand at the t-th moment, C NH,in,t is the influent ammonia nitrogen concentration at the t-th moment, and t is the number of the moment; Calculate the water quality index change coefficient based on the difference in water quality indices, and construct a water quality index change sequence: , where ε is the water quality index change coefficient, and | | represents the absolute value operation; S6. Predict the water quality treatment fitness score based on the water quality prediction model according to the residence time deviation sequence, the aeration volume deviation sequence, and the water quality index change sequence; The water quality prediction model includes: a time deviation processing channel, an aeration volume deviation processing channel, a water quality index change processing channel, and an output layer; The input end of the time deviation processing channel is used to input the residence time deviation sequence to obtain a time influence value; the input end of the aeration volume deviation processing channel is used to input the aeration volume deviation sequence to obtain an aeration volume influence value; the input end of the water quality index change processing channel is used to input the water quality index change sequence to obtain a water quality index influence value; the output layer is used to weight the time influence value, the aeration volume influence value, and the water quality index influence value to obtain the water quality treatment fitness score; The time deviation processing channel, the aeration volume deviation processing channel, and the water quality index change processing channel all include, connected in sequence: a max pooling layer, a significant value enhancement layer, and a fully connected layer. Among them, the max pooling layer is a one-dimensional pooling layer, the pooling window size is 4, and the stride is 4; The expression of the significant value enhancement layer is as follows: , where y i is the i-th enhanced significant value output by the significant value enhancement layer, x i is the i-th significant value output by the max pooling layer, and e is the natural constant.

2. The sewage treatment water quality prediction method according to claim 1, wherein The reference time in S3 includes: the minimum residence time T min , the maximum residence time T max and the standard residence time T s , and S3 includes the following sub-steps: S31. When (T min + T t ) / 2 ≤ T t ≤ (T max + T t ), the time deviation coefficient assigned to the corresponding sewage retention time T t is 0; S32. At T min ≤T t <(T min +T s ) / 2, calculate the time deviation coefficient corresponding to the sewage retention time T t : , where θ is the time deviation coefficient; S33. When (T max + T s ) / 2 < T t ≤ T max , calculate the time deviation coefficient corresponding to the sewage retention time T t : ; S34. At T t <T min When, calculate the time deviation coefficient corresponding to the sewage residence time T t : ; S35. At T t > T max When, calculate the time deviation coefficient corresponding to the sewage retention time T t : ; S36. Arrange each time deviation coefficient in the order of occurrence according to the sewage residence time T t to obtain a residence time deviation sequence.

3. The sewage treatment water quality prediction method according to claim 1, characterized in that The S4 includes the following sub-steps: S41. Calculate the required aeration volume for the influent water quality indicators at each moment to obtain a required aeration volume sequence; S42. Obtain a comparison aeration volume sequence according to the aeration volume provided at each historical moment; S43. Subtract the comparison aeration volume sequence from the required aeration volume sequence to obtain an aeration volume difference sequence; S44. Calculate the aeration volume deviation coefficient according to each aeration volume difference in the aeration volume difference sequence, and construct an aeration volume deviation sequence.

4. A sewage treatment water quality prediction system, which is implemented based on the sewage treatment water quality prediction method according to any one of claims 1 to 3, and is characterized in that It includes: A collection unit, a sewage residence time acquisition unit, a first sequence construction unit, a second sequence construction unit, a third sequence construction unit, and a prediction unit; The collection unit is used to collect the sewage influent flow rate and influent water quality indicators at each moment; The sewage residence time acquisition unit is used to obtain multiple sewage residence times in the biochemical pool according to the sewage influent flow rate at each moment; The first sequence construction unit is used to construct a residence time deviation sequence according to the gap between the multiple sewage residence times and the reference time; The second sequence construction unit is used to construct an aeration volume deviation sequence according to the gap between the aeration volume provided at the historical moment and the required aeration volume; The third sequence construction unit is used to construct a water quality index change sequence according to the influent water quality indicators at each moment; The prediction unit is used to predict the water quality treatment fitness score based on the water quality prediction model according to the residence time deviation sequence, the aeration volume deviation sequence, and the water quality index change sequence.

Images