Tests of Inflatable Structure Shape Control Using Genetic Algorithm and Neural Network
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Inflatable space structures need to maintain in a desired shape in space in order to achieve satisfactory performance. The active shape control technique has shown its advantages in solving this problem. Due to strong non-linear properties of the inflatable structures, it is a challenging task to model the inflatable structure properties and to find optimal control output. In this paper, a scheme is proposed based on genetic algorithm and neural network, which is then verified on the shape control of a small size membrane structure. The membrane to be controlled is a 200mm × 300mm rectangular Kapton membrane, pulled by two tensions along each edge. Different combinations of the tensions produce various wrinkles on the membrane. A neural network model is developed to map boundary stretching tensions and space environment to membrane flatness, and then is used to estimate the membrane flatness. The genetic algorithm is utilized to search the best tension combinations from the neural network model to minimize the membrane wrinkle amplitude. An active control system is developed and tests are performed. The results show that genetic algorithm works very well in optimizing the tensions and neural network is effective to estimate the flatness of the membrane. Nomenclature xi = neural network input vij, wjk = neural network weights βj, ρk = neural network thresholds Yk = neural network output γ j = combining function output zj = activation function output n = number of neurons in input layer p = number of neurons in hidden layer m = number of neurons in output layer I I.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it