Method and device for determining the charge of ink droplet sequences of a code jet

Abstract

The application provides a method and device for determining the charge amount of an ink droplet sequence of a code jet instrument, which comprises: obtaining the theoretical charge amount of a current ink droplet sequence based on the expected printing position of the current ink droplet sequence jetted by the code jet instrument and the force of the deflection electric field of the code jet instrument on the current ink droplet sequence; inputting the expected printing position into a BiLSTM network to obtain the charge amount compensation value of the current ink droplet sequence output by the BiLSTM network; obtaining the actual charge amount of the current ink droplet sequence based on the theoretical charge amount and the charge amount compensation value of the current ink droplet sequence; the BiLSTM network is obtained by training with the actual printing position of an ink droplet sequence sample as a sample and the difference between the theoretical charge amount and the actual charge amount of the ink droplet sequence sample as a label, and the theoretical charge amount of the ink droplet sequence sample is obtained based on the actual printing position. The application accurately calculates the charge amount of the current ink droplet sequence, so that the actual printing position of the current ink droplet sequence is close to the expected printing position.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and apparatus for determining the charge of ink droplet sequences in an inkjet printer. Background Technology

[0002] Theoretically, the printing position (landing point) of the ink droplets ejected by the inkjet printer depends only on the charge of the ink droplets and the voltage of the deflection electric field. Current technology only considers the effect of the deflection electric field on the charged ink droplets, and calculates the theoretically required charge to be applied to the ink droplets from the desired printing position under a certain deflection voltage.

[0003] However, in reality, charged ink droplets are affected by air resistance, gravity, and the forces exerted by other charged ink droplets as they fly through the air. This results in a non-linear relationship between the final printing position (deflection distance) and the charge value of the charging electrode. Applying a theoretical charge to the ink droplet will cause a deviation between the actual and desired printing position. Summary of the Invention

[0004] This invention provides a method and apparatus for determining the charge of ink droplet sequences in an inkjet printer, which solves the problem of inaccurate determination of ink droplet charge in the prior art and improves the accuracy of ink droplet charge calculation.

[0005] This invention provides a method for determining the charge level of ink droplet sequences in an inkjet printer, comprising:

[0006] Based on the desired printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence, the theoretical charge of the current ink droplet sequence is obtained.

[0007] The desired printing position is input into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network.

[0008] Based on the theoretical charge and charge compensation value of the current ink droplet sequence, the actual charge of the current ink droplet sequence is obtained;

[0009] The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0010] According to the present invention, a method for determining the charge level of an ink droplet sequence in an inkjet printer includes inputting the desired printing position into a BiLSTM network to obtain the charge level compensation value of the current ink droplet sequence output by the BiLSTM network, comprising:

[0011] The desired printing position of the current ink droplet sequence, as well as the temperature and humidity of the environment in which the current ink droplet sequence is located, are input into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network.

[0012] According to the present invention, a method for determining the charge level of an ink droplet sequence in an inkjet printer includes inputting the desired printing position into a BiLSTM network to obtain the charge level compensation value of the current ink droplet sequence output by the BiLSTM network, comprising:

[0013] The desired printing position is input into the forward LSTM and backward LSTM in the BiLSTM network to obtain the first feature output by the forward LSTM and the second feature output by the backward LSTM;

[0014] After concatenating the first feature and the second feature, the sequence passes through the first linear layer, the activation function layer, and the second linear layer to obtain the charge compensation value of the current ink droplet sequence output by the second linear layer.

[0015] According to a method for determining the charge level of an ink droplet sequence in an inkjet printer provided by the present invention, before inputting the desired printing position into a BiLSTM network to obtain the charge level compensation value of the current ink droplet sequence output by the BiLSTM network, the method further includes:

[0016] The actual printing position of the ink droplet sequence sample is input into the BiLSTM network to obtain the charge compensation value of the ink droplet sequence sample output by the BiLSTM network.

[0017] The mean square error between the charge compensation value of the ink droplet sequence samples and the difference is used as the objective function to train the BiLSTM network.

[0018] According to a method for determining the charge level of an ink droplet sequence in an inkjet printer provided by the present invention, before inputting the desired printing position into a BiLSTM network to obtain the charge level compensation value of the current ink droplet sequence output by the BiLSTM network, the method further includes:

[0019] The BiLSTM network was trained using the stochastic gradient descent method.

[0020] According to a method for determining the charge charge of an ink droplet sequence in an inkjet printer provided by the present invention, after obtaining the actual charge charge of the current ink droplet sequence based on the theoretical charge charge and the charge charge compensation value, the method further includes:

[0021] The charging voltage of the inkjet printer's charging electrode is determined based on the actual charge carried by the current ink droplet sequence.

[0022] The present invention also provides a device for determining the charge of ink droplet sequence in an inkjet printer, comprising:

[0023] The acquisition module is used to acquire the theoretical charge of the current ink droplet sequence based on the expected printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence.

[0024] The prediction module is used to input the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network.

[0025] The calculation module is used to obtain the actual charge of the current ink droplet sequence based on the theoretical charge and charge compensation value of the current ink droplet sequence;

[0026] The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0027] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the ink droplet sequence charge determination method of any of the above-described methods.

[0028] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the ink droplet sequence charge determination method of the inkjet printer as described above.

[0029] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the ink droplet sequence charge determination method of the inkjet printer as described above.

[0030] The present invention provides a method and apparatus for determining the charge of ink droplet sequences in an inkjet printer. By comprehensively considering factors affecting the printing position of ink droplets, and given the desired printing position of the current ink droplet sequence, the present invention uses a BiLSTM network to predict the charge compensation value of the current ink droplet sequence. The charge compensation value is then used to compensate for the theoretical charge of the current ink droplet sequence obtained by considering only the deflection electric field, thereby accurately calculating the charge of the current ink droplet sequence and making the actual printing position of the current ink droplet sequence close to the desired printing position. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the working principle of the inkjet printer in the method for determining the charge of ink droplet sequence provided by the present invention;

[0033] Figure 2 This is a schematic diagram of the LSTM network structure in the ink droplet sequence charge determination method for inkjet printers provided by the present invention;

[0034] Figure 3 This is a schematic diagram comparing the structures of LSTM and RNN in the method for determining the charge of ink droplet sequence in an inkjet printer provided by the present invention;

[0035] Figure 4 This is a schematic diagram of the calculation process of the LSTM network unit in the method for determining the charge of ink droplet sequence in an inkjet printer provided by the present invention.

[0036] Figure 5 This is a flowchart illustrating the method for determining the charge of ink droplet sequences in an inkjet printer provided by the present invention.

[0037] Figure 6 This is a schematic diagram of the process for determining the charge compensation value of an ink droplet sequence using a BiLSTM network in the method for determining the charge of an ink droplet sequence provided by the present invention.

[0038] Figure 7 This is a schematic diagram of the device for determining the charge of ink droplet sequence in an inkjet printer provided by the present invention;

[0039] Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0041] Figure 1 This is a schematic diagram illustrating the working principle of the inkjet printer in the method for determining the charge of ink droplet sequences provided by this invention. Figure 1As shown, conductive ink enters the ink chamber under pressure and is then ejected from the nozzle. As the ink passes through the nozzle, it is broken into a continuous series of equally spaced and identical-sized droplets by the action of a piezoelectric crystal. The ejected ink stream travels downwards through a charging electrode, where the droplets separate from the ink stream.

[0042] A specific voltage is applied to the charging electrode. When an ink droplet separates from the conductive ink line, it instantly acquires a negative charge proportional to the applied voltage. The voltage frequency of the charging electrode is exactly the same as the frequency at which the ink droplet breaks apart, ensuring that each droplet is accurately charged with the predetermined negative charge. The droplet then continues its downward trajectory, passing through an electric field created by two deflecting plates with a voltage of approximately 10 kV. Charged droplets are deflected upon passing through this electric field, the degree of deflection depending on the amount of charge they carry. Uncharged droplets do not deflect and continue flying downwards, flowing into the recycling tank and eventually returning to the ink cartridge for reuse. Charged and deflected droplets fall at a specific speed and angle onto objects passing under the printhead. By changing the charge carried by the droplets, different patterns can be generated, including printing letters, numbers, and graphics.

[0043] The force exerted by the deflecting electric field on the charged ink droplet is the main factor affecting the droplet's printing position. Assuming the deflection voltage is U, the charge of the charged ink droplet is q, the electric field strength is E, and the distance between the two electrode plates of the deflecting electric field is d, then the electric force on the charged ink droplet is q·E=q·U / d.

[0044] The charged ink droplet is also subject to the force of neighboring charged ink droplets, which can be calculated using Coulomb's law. Where q1 and q2 are the charges of the charged ink droplets, r is the distance between the ink droplets, and k is the Coulomb constant.

[0045] For the current ink droplet, if the charge of the preceding and following ink droplets is exactly the same, the electric forces they experience cancel each other out. However, due to printing requirements, to achieve different printing positions, the preceding and following ink droplets need to carry different charges. Therefore, the electric forces exerted by the preceding and following ink droplets on the current ink droplet cannot cancel each other out. Thus, it is necessary to calculate each force individually according to Coulomb's law, and finally synthesize them according to the vector composition rule of forces. The nature of these electric forces depends on the charge of the preceding and following ink droplets and the current ink droplet, making modeling quite complex.

[0046] Meanwhile, temperature and humidity affect the movement of ink droplets. Inevitably, ink droplets are also affected by gravity and air resistance in a deflecting electric field.

[0047] In summary, the relationship between the printing position of the ink droplet sequence and the charge of the ink droplet sequence is complex, requiring consideration of how to calculate the required charge applied to the ink droplet sequence starting from its printing position. Therefore, how to model the relationship between the charge of the ink droplets and the printing position is the problem addressed in this invention.

[0048] Figure 2 This is a schematic diagram of the LSTM network structure in the ink droplet sequence charge determination method for inkjet printers provided by this invention. Figure 2 As shown, LSTM (Long Short-Term Memory), a member of the recurrent neural network family, is an improvement on RNN (Recurrent Neural Network). LSTM addresses the long-term dependency problem commonly found in general recurrent neural networks, effectively transmitting and representing information over long time sequences without forgetting useful information from long ago. Simultaneously, LSTM can also solve the vanishing or exploding gradient problems in RNNs. Compared to RNNs, the innovation of LSTM lies in introducing two transmission states: the long-term memory cell state and the short-term memory hidden state, which are passed from the current neuron to the next. In contrast, RNNs only have one hidden state passed between neurons. Figure 3 It demonstrates the differences between the two.

[0049] LSTM introduces three gate signals: the forget gate, the remember gate, and the output gate. The forget gate and the remember gate are responsible for forgetting unimportant information and remembering important information, respectively, while the output gate determines the information that will be output.

[0050] Figure 4 This is a schematic diagram of the calculation process of the LSTM network unit in the ink droplet sequence charge determination method for inkjet printers provided by this invention. Figure 4 As shown, firstly, the hidden state output from the previous time step... t-1 and the input x at the current time t By concatenating the components and applying them to three weight matrices and a sigmoid activation function, the forget gate z is obtained. f Memory gate z i and output gate z o The values ​​of z are all between 0 and 1; the effective input signal z is generated by the weight matrix and the tanh hyperbolic tangent activation function, and the values ​​of z are between -1 and 1.

[0051] First, forget the gate z f and the cell state of the previous moment t-1 Perform positional multiplication to selectively forget unimportant information in the cell state; use a memory gate z-axis. i Multiplying the input by z in pairs allows for selective memorization of important information; then, the results of the two pairwise multiplications are summed as a matrix to obtain the cell state at the current time step. tThe hidden state at the current moment. t This means c t After passing through the tanh activation function, it is then compared with the output gating signal z. o The result is obtained by performing positional multiplication. Finally, the output y at the current time step... t It is h t It is obtained after passing through the weight matrix and the sigmoid activation function.

[0052] BiLSTM is essentially a regular unidirectional LSTM with an additional backward LSTM. The forward and backward LSTMs are then concatenated, allowing the later data to utilize information from the earlier data, and vice versa. The backward LSTM refers to the LSTM unit outputting c from the next time step. t+1 h t+1 The data is passed forward to the current time step, and otherwise behaves the same as the forward LSTM. Another way to understand this is to reverse the time sequence and input it into the forward LSTM network, achieving the same effect as the backward LSTM. During concatenation, the hidden states output at the same time step from both the forward and backward LSTMs are concatenated.

[0053] The following is combined with Figure 5 The present invention describes a method for determining the charge charge of an ink droplet sequence in an inkjet printer, comprising:

[0054] Step 501: Based on the desired printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence, obtain the theoretical charge of the current ink droplet sequence.

[0055] The desired printing position of the current ink droplet sequence can be determined based on the position of the printed content of the current ink droplet sequence on the substrate surface.

[0056] In an inkjet printer, charged ink droplets are deflected by a deflecting electric field, leaving the field at a certain speed and angle, and thus reaching the surface of the substrate to form the printing position. The printing position is affected not only by the force exerted by the deflecting electric field on the charged ink droplets, but also by the interaction force between the charged ink droplets, temperature, humidity, and other factors.

[0057] By considering only the force exerted by the deflection electric field on the current ink droplet sequence, and by working backward from the expected printing position of the current ink droplet sequence, the theoretical charge of the current ink droplet sequence can be calculated, that is, the theoretical charge that needs to be applied to the current ink droplet sequence.

[0058] Step 502: Input the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network;

[0059] Due to factors such as the interaction force between successive ink droplets, gravity, air resistance, temperature, and humidity, if a theoretical charge is applied to the current ink droplet sequence, there will be a deviation between the actual and desired printing positions. Therefore, to position the current ink droplet sequence at the desired printing position, there is a discrepancy between the theoretically required charge and the actual charge required. This discrepancy serves as a charge compensation value for the current ink droplet sequence. This charge compensation value can be obtained using a neural network.

[0060] Considering the temporal influence of the preceding and following ink droplets in the ink droplet stream, a BiLSTM (Bi-directional Long Short-Term Memory) network model can be used. The desired ink droplet printing position sequence is taken as the model input, and the output is the charge compensation value for the current ink droplet sequence.

[0061] The reason for choosing BiLSTM as the network model is that the actual charge that places the current ink droplet sequence at the desired printing position is affected not only by the previous printing position but also by the subsequent printing position.

[0062] Step 503: Based on the theoretical charge and charge compensation value of the current ink droplet sequence, obtain the actual charge of the current ink droplet sequence;

[0063] Add the charge compensation value of the current ink droplet sequence output by the BiLSTM network to the theoretical charge value to obtain the actual charge value of the current ink droplet sequence, which is the actual charge value that should be applied to the ink droplet sequence.

[0064] The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0065] Before using the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence, the BiLSTM network is trained.

[0066] In the experiment, the voltage value of the charging electrode of the inkjet printer was varied, and proportional conversion was performed to ensure that the ink droplet sequence samples carried different actual charges, resulting in a sample length ranging from 3 to 10. After passing through a deflection electric field, the ink droplet sequence samples were deflected, leaving the field at a certain angle and speed, and reaching the substrate surface. The substrate was placed on a conveyor belt and moved with it at a certain speed.

[0067] During the experiment, the printing height and conveyor belt speed were kept constant. The position where the ink droplet sequence sample reached the surface of the printed object needed to be recorded, i.e., the actual printing position. This actual printing position required observation using an industrial microscope with a magnification of several hundred times. To avoid the influence of other factors, the voltage of the deflection electric field and the acceleration due to gravity should be kept constant during the experiment.

[0068] The recorded experimental data includes the actual charge and actual printing position of the ink droplet sequence samples. Based on the actual printing position of the ink droplet sequence, and considering only the force exerted by the deflection electric field on the ink droplet sequence samples, the theoretical charge of the ink droplet sequence samples can be calculated. Combining the actual charge recorded in the experiment, the difference (compensation) between the two is calculated, and this difference is used as the label.

[0069] The desired printing position of the ink droplet sequence samples is used as input to the BiLSTM network to obtain the charge compensation value for the ink droplet sequence samples. The desired printing position of the ink droplet sequence samples and the label are divided into training and test sets according to a preset ratio, such as 8:2, and then the BiLSTM network is trained.

[0070] Before training, the training data needs to be normalized. Since the lengths of the ink droplet sequence samples are inconsistent, the pad_sequence method can be used to pad the unequal-length expected printing position sequences to the same length, which is the longest length of the ink droplet sequence in the training data. This allows the padded, equal-length sequence to be input into the BiLSTM network.

[0071] This embodiment comprehensively considers the factors affecting the printing position of ink droplets. Given the desired printing position of the current ink droplet sequence, it uses a BiLSTM network to predict the charge compensation value of the current ink droplet sequence. The charge compensation value is used to compensate for the theoretical charge of the current ink droplet sequence obtained by only considering the deflection electric field, thereby accurately calculating the charge of the current ink droplet sequence and making the actual printing position of the current ink droplet sequence close to the desired printing position.

[0072] Based on the above embodiments, the step of inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network includes:

[0073] The desired printing position of the current ink droplet sequence, as well as the temperature and humidity of the environment in which the current ink droplet sequence is located, are input into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network.

[0074] This embodiment takes into account the effects of temperature and humidity, and uses the desired printing position of the current ink droplet sequence, the temperature and humidity of the environment as inputs to the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence.

[0075] During the training of the BiLSTM network, the charging voltage was varied under different temperature and humidity conditions, and the experiment was repeated multiple times, recording the actual printing positions. The desired printing positions of the ink droplet sequence samples, along with the temperature and humidity of the environment, were used as inputs to the BiLSTM network to obtain the charge compensation values ​​for the ink droplet sequence samples.

[0076] This embodiment takes into account the influence of temperature and humidity, and also uses the temperature and humidity of the current ink droplet sequence's environment as input to the BiLSTM network, which affects the compensation value of the charge of the ink droplet sequence output by the BiLSTM network, thereby accurately calculating the charge of the current ink droplet sequence, so that the actual printing position of the current ink droplet sequence is close to the desired printing position.

[0077] Based on the above embodiments, the step of inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network includes:

[0078] The desired printing position is input into the forward LSTM and backward LSTM in the BiLSTM network to obtain the first feature output by the forward LSTM and the second feature output by the backward LSTM;

[0079] After concatenating the first feature and the second feature, the sequence passes through the first linear layer, the activation function layer, and the second linear layer to obtain the charge compensation value of the current ink droplet sequence output by the second linear layer.

[0080] like Figure 6 As shown, the input feature dimension `input_size` of the BiLSTM network is 1, which represents the desired printing position. It can also be 3, including the desired printing position of the current ink droplet sequence, the ambient temperature, and humidity. The hidden state dimension `hidden_size` can be set to 100, and the number of hidden layers `num_layers` can be set to 2.

[0081] The output of the BiLSTM network is successively input into a linear layer, a tanh activation function layer, and another linear layer to ensure that the output dimension is 1, which is the charge compensation value for the current ink droplet sequence.

[0082] This embodiment comprehensively considers the factors affecting the printing position of ink droplets. Given the desired printing position of the current ink droplet sequence, it uses a BiLSTM network to predict the charge compensation value of the current ink droplet sequence. The charge compensation value is used to compensate for the theoretical charge of the current ink droplet sequence obtained by only considering the deflection electric field, thereby accurately calculating the charge of the current ink droplet sequence and making the actual printing position of the current ink droplet sequence close to the desired printing position.

[0083] Based on the above embodiments, this embodiment further includes the following step before inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network:

[0084] The actual printing position of the ink droplet sequence sample is input into the BiLSTM network to obtain the charge compensation value of the ink droplet sequence sample output by the BiLSTM network.

[0085] The BiLSTM network is trained using the mean square error (MSE) between the charge compensation value of the ink droplet sequence samples and the difference as the objective function.

[0086] During the training of the BiLSTM network, the weight parameters of the BiLSTM network are adjusted to minimize the objective function, thereby accurately calculating the charge of the current ink droplet sequence and making the actual printing position of the current ink droplet sequence close to the desired printing position.

[0087] Based on the above embodiments, this embodiment further includes the following step before inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network:

[0088] The BiLSTM network was trained using the Stochastic Gradient Descent (SGD) method.

[0089] Based on the above embodiments, this embodiment, after obtaining the actual charge of the current ink droplet sequence based on the theoretical charge and charge compensation value of the current ink droplet sequence, further includes:

[0090] The charging voltage of the inkjet printer's charging electrode is determined based on the actual charge carried by the current ink droplet sequence.

[0091] Based on the proportional relationship between the charging voltage and the actual charge of the current droplet sequence, the charging voltage of the charging electrode can be determined after determining the actual charge of the current droplet sequence. Applying a determined charging voltage to the charging electrode can bring the actual printing position of the current droplet sequence closer to the desired printing position.

[0092] The inkjet printer droplet sequence charge determination device provided by the present invention is described below. The inkjet printer droplet sequence charge determination device described below can be referred to in correspondence with the inkjet printer droplet sequence charge determination method described above.

[0093] like Figure 7As shown, the device includes an acquisition module 701, a prediction module 702, and a calculation module 703, wherein:

[0094] The acquisition module 701 is used to acquire the theoretical charge of the current ink droplet sequence based on the expected printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence.

[0095] The prediction module 702 is used to input the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network;

[0096] The calculation module 703 is used to obtain the actual charge of the current ink droplet sequence based on the theoretical charge and charge compensation value of the current ink droplet sequence;

[0097] The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0098] This invention comprehensively considers the factors affecting the printing position of ink droplets. Given the desired printing position of the current ink droplet sequence, it uses a BiLSTM network to predict the charge compensation value of the current ink droplet sequence. The charge compensation value is used to compensate for the theoretical charge of the current ink droplet sequence obtained by only considering the deflection electric field, thereby accurately calculating the charge of the current ink droplet sequence and making the actual printing position of the current ink droplet sequence close to the desired printing position.

[0099] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call logic instructions in the memory 830 to execute a method for determining the charge of an ink droplet sequence. This method includes: obtaining the theoretical charge of the current ink droplet sequence based on the desired printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence; inputting the desired printing position into a BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network; and obtaining the actual charge of the current ink droplet sequence based on the theoretical charge and the charge compensation value. The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as a sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as a label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0100] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0101] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the ink droplet sequence charge determination method provided by the above methods. The method includes: obtaining the theoretical charge of the current ink droplet sequence based on the expected printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence; inputting the expected printing position into a BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network; and obtaining the actual charge of the current ink droplet sequence based on the theoretical charge and the charge compensation value. The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as a sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as a label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0102] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for determining the charge of an ink droplet sequence provided by the above methods. The method includes: obtaining the theoretical charge of the current ink droplet sequence based on the desired printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence; inputting the desired printing position into a BiLSTM network to obtain a charge compensation value for the current ink droplet sequence output by the BiLSTM network; and obtaining the actual charge of the current ink droplet sequence based on the theoretical charge and the charge compensation value. The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as a sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as a label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position.

[0103] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining the charge of an ink droplet sequence in an inkjet printer, characterized in that, include: Based on the desired printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence, the theoretical charge of the current ink droplet sequence is obtained. The desired printing position is input into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network. Based on the theoretical charge and charge compensation value of the current ink droplet sequence, the actual charge of the current ink droplet sequence is obtained; The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position. The step of inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network includes: The desired printing position is input into the forward LSTM and backward LSTM in the BiLSTM network to obtain the first feature output by the forward LSTM and the second feature output by the backward LSTM; After concatenating the first feature and the second feature, the sequence passes through the first linear layer, the activation function layer, and the second linear layer to obtain the charge compensation value of the current ink droplet sequence output by the second linear layer.

2. The method for determining the charge of ink droplet sequence in an inkjet printer according to claim 1, characterized in that, The step of inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network includes: The desired printing position of the current ink droplet sequence, as well as the temperature and humidity of the environment in which the current ink droplet sequence is located, are input into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network.

3. The method for determining the charge of ink droplet sequence in an inkjet printer according to claim 1 or 2, characterized in that, Before inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network, the method further includes: The actual printing position of the ink droplet sequence sample is input into the BiLSTM network to obtain the charge compensation value of the ink droplet sequence sample output by the BiLSTM network. The mean square error between the charge compensation value of the ink droplet sequence samples and the difference is used as the objective function to train the BiLSTM network.

4. The method for determining the charge of ink droplet sequence in an inkjet printer according to claim 3, characterized in that, Before inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network, the method further includes: The BiLSTM network was trained using the stochastic gradient descent method.

5. The method for determining the charge of ink droplet sequence in an inkjet printer according to claim 1 or 2, characterized in that, After obtaining the actual charge of the current ink droplet sequence based on the theoretical charge and charge compensation value of the current ink droplet sequence, the method further includes: The charging voltage of the inkjet printer's charging electrode is determined based on the actual charge carried by the current ink droplet sequence.

6. A device for determining the charge level of ink droplet sequence in an inkjet printer, characterized in that, include: The acquisition module is used to acquire the theoretical charge of the current ink droplet sequence based on the expected printing position of the current ink droplet sequence ejected by the inkjet printer and the force exerted by the deflection electric field of the inkjet printer on the current ink droplet sequence. The prediction module is used to input the desired printing position into the BiLSTM network to obtain the charge compensation value of the current ink droplet sequence output by the BiLSTM network. The calculation module is used to obtain the actual charge of the current ink droplet sequence based on the theoretical charge and charge compensation value of the current ink droplet sequence; The BiLSTM network is trained using the actual printing position of the ink droplet sequence sample as the sample and the difference between the theoretical charge and the actual charge of the ink droplet sequence sample as the label. The theoretical charge of the ink droplet sequence sample is obtained based on the actual printing position. The step of inputting the desired printing position into the BiLSTM network to obtain the charge compensation value of the current droplet sequence output by the BiLSTM network includes: The desired printing position is input into the forward LSTM and backward LSTM in the BiLSTM network to obtain the first feature output by the forward LSTM and the second feature output by the backward LSTM; After concatenating the first feature and the second feature, the sequence passes through the first linear layer, the activation function layer, and the second linear layer to obtain the charge compensation value of the current ink droplet sequence output by the second linear layer.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for determining the charge of ink droplet sequence in an inkjet printer as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for determining the charge of ink droplet sequence in an inkjet printer as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for determining the charge of ink droplet sequence in an inkjet printer as described in any one of claims 1 to 5.

Images